|
»
»
can't read sequential files
|
View:
New views
12 Messages
—
Rating Filter:
Alert me
|
 |
can't read sequential files
by zirtik
::
Rate this Message:
Hi, I'm using cygwin and windows XP together with Eclipse IDE and CDT. I have a following piece of code:
.... int i; fp = fopen ("phi.txt","r"); for( i = 0; i < 51; i++ ) { fscanf(fp, "%d\n", &original_phi[i]); } ... and when I try to compile it, it compiles well, but when I try to run it I get the following error message: 655 [main] Genetics 3012 _cygtls::handle_exceptions: Exception: STATUS_ACCESS_VIOLATION 2540 [main] Genetics 3012 open_stackdumpfile: Dumping stack trace to Genetics.exe.stackdump I added the following line to check if fp is always NULL: if (fp==NULL) { printf("error, NULL pointer!\n"); return(1); } and it is always NULL. I put the "phi.txt" in the same directory as the executable so it is in: ....\ProjectFolder\Debug folder as well as in ....\ProjectFolder\src folder. I still keep getting the same error. I added the current directory "." to the PATH but it didn't help. This code used to run on another PC with Eclipse and cygwin again but this is the first time I am having problems with it on a different machine. Any help would be greatly appreciated. Thanks. |
|
|
Re: can't read sequential files
by jim marshall
::
Rate this Message:
zirtik wrote:
> Hi, I'm using cygwin and windows XP together with Eclipse IDE and CDT. I have > a following piece of code: > > .... > > int i; > fp = fopen ("phi.txt","r"); > > for( i = 0; i < 51; i++ ) { > fscanf(fp, "%d\n", &original_phi[i]); > } > > ... > > and when I try to compile it, it compiles well, but when I try to run it I > get the following error message: > > 655 [main] Genetics 3012 _cygtls::handle_exceptions: Exception: > STATUS_ACCESS_VIOLATION > 2540 [main] Genetics 3012 open_stackdumpfile: Dumping stack trace to > Genetics.exe.stackdump > > I added the following line to check if fp is always NULL: > > if (fp==NULL) > { > printf("error, NULL pointer!\n"); > return(1); > } > > and it is always NULL. I put the "phi.txt" in the same directory as the > executable so it is in: > > ....\ProjectFolder\Debug > > folder as well as in > > ....\ProjectFolder\src > > folder. I still keep getting the same error. I added the current directory > "." to the PATH but it didn't help. This code used to run on another PC with > Eclipse and cygwin again but this is the first time I am having problems > with it on a different machine. Any help would be greatly appreciated. > > Thanks. > > > (see perror, or strerror functions) to see what the error is. -- 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: can't read sequential files
by Brian Dessent
::
Rate this Message:
zirtik wrote:
> and when I try to compile it, it compiles well, but when I try to run it I > get the following error message: > > 655 [main] Genetics 3012 _cygtls::handle_exceptions: Exception: > STATUS_ACCESS_VIOLATION > 2540 [main] Genetics 3012 open_stackdumpfile: Dumping stack trace to > Genetics.exe.stackdump There's a bug in your program that causes a segfault. We can't know what it is because you only posted a snippet. We need a fully standalone (compilable) testcase that exibits the problem. Otherwise, how are we supposed to know what you're doing? 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: can't read sequential files
by Alberto Luaces
::
Rate this Message:
Maybe unrelated, but:
shouldn't this > fp = fopen ("phi.txt","r"); be fp = fopen ("phi.txt","rt"); ? -- 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: can't read sequential files
by Dave Korn
::
Rate this Message:
On 31 October 2007 00:11, zirtik wrote:
> int i; > fp = fopen ("phi.txt","r"); > > for( i = 0; i < 51; i++ ) { > fscanf(fp, "%d\n", &original_phi[i]); > } > 655 [main] Genetics 3012 _cygtls::handle_exceptions: Exception: > STATUS_ACCESS_VIOLATION > 2540 [main] Genetics 3012 open_stackdumpfile: Dumping stack trace to > Genetics.exe.stackdump > if (fp==NULL) > { > printf("error, NULL pointer!\n"); > return(1); > } > > and it is always NULL. You should also print out the value of 'errno' here, it's the global variable that holds the code for the last error from C library functions. > I put the "phi.txt" in the same directory as the > executable so it is in: > > ....\ProjectFolder\Debug > > folder as well as in > > ....\ProjectFolder\src > > folder. I still keep getting the same error. It would tell us a lot if we knew whether errno was equal to ENOENT or EACCES. > I added the current directory > "." to the PATH but it didn't help. Nope, it wouldn't make a difference. PATH is only used by the shell when searching for commands; it isn't used when you're calling fopen. For fopen, you must provide the exact path to the file, or the file must be in the cwd (current directory) when your program is running. > This code used to run on another PC with > Eclipse and cygwin again but this is the first time I am having problems > with it on a different machine. Any help would be greatly appreciated. I don't use eclipse myself, but I'd guess there's a setting somewhere for what the current directory should be when it launches the executable. Make sure your text file is in that directory, or change the setting so it points at the debug/release folder where you already put the text file. 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: can't read sequential files
by Larry Hall (Cygwin)
::
Rate this Message:
Alberto Luaces wrote:
> Maybe unrelated, but: > > shouldn't this > >> fp = fopen ("phi.txt","r"); > > be > > fp = fopen ("phi.txt","rt"); > > ? It's not required, no. -- Larry Hall http://www.rfk.com RFK Partners, Inc. (508) 893-9779 - RFK Office 216 Dalton Rd. (508) 893-9889 - FAX Holliston, MA 01746 _____________________________________________________________________ A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting annoying in email? -- 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: can't read sequential files
by zirtik
::
Rate this Message:
After adding the line:
if (fp==NULL) { printf("error, NULL pointer!\n"); return(1); } and then rebuilding the code, everything worked. But it's strange that if I delete that code segment and never check whether the fp pointer is NULL or not, I always get a segmentation fault. Can this be some kind of an optimization problem? I don't know why it happens. Thank you for the comments.
|
|
|
RE: can't read sequential files
by Dave Korn
::
Rate this Message:
On 01 November 2007 06:43, zirtik wrote:
> After adding the line: > > > if (fp==NULL) > { > printf("error, NULL pointer!\n"); > return(1); > } > > and then rebuilding the code, everything worked. But it's strange that if I > delete that code segment and never check whether the fp pointer is NULL or > not, I always get a segmentation fault. Can this be some kind of an > optimization problem? I don't know why it happens. Thank you for the > comments. A NULL pointer is never valid in C, and a segfault is what you get if you try to make use of one (by dereferencing it). You get a NULL pointer back from fopen when it fails; a lot of library routines do this to indicate failure, because any other value could be a valid pointer. The other library routines, such as fread and fwrite, will assume that you have done your error checking and won't be passing them a NULL pointer, so they won't bother to check what file pointer you pass them, they'll just go ahead and try and use it. So if you get a NULL pointer back from fopen and you don't check for it, your code carries on and passes that same pointer to fread, which tries to use it as if it pointed to a real FILE object, and crashes. The comp.lang.c FAQ has an entire section on NULL pointer, section 5. It should be available at http://c-faq.com/null/index.html but the website seems to be temporarily down right now; there's also a copy at faqs.org: http://www.faqs.org/faqs/C-faq/faq/ 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: can't read sequential files
by Lewis Hyatt
::
Rate this Message:
Dave Korn wrote:
> On 01 November 2007 06:43, zirtik wrote: > >> After adding the line: >> >> >> if (fp==NULL) >> { >> printf("error, NULL pointer!\n"); >> return(1); >> } >> >> and then rebuilding the code, everything worked. But it's strange that if I >> delete that code segment and never check whether the fp pointer is NULL or >> not, I always get a segmentation fault. Can this be some kind of an >> optimization problem? I don't know why it happens. Thank you for the >> comments. > > A NULL pointer is never valid in C, and a segfault is what you get if you > try to make use of one (by dereferencing it). You get a NULL pointer back > from fopen when it fails; a lot of library routines do this to indicate > failure, because any other value could be a valid pointer. The other library > routines, such as fread and fwrite, will assume that you have done your error > checking and won't be passing them a NULL pointer, so they won't bother to > check what file pointer you pass them, they'll just go ahead and try and use > it. So if you get a NULL pointer back from fopen and you don't check for it, > your code carries on and passes that same pointer to fread, which tries to use > it as if it pointed to a real FILE object, and crashes. > > The comp.lang.c FAQ has an entire section on NULL pointer, section 5. It > should be available at > http://c-faq.com/null/index.html > but the website seems to be temporarily down right now; there's also a copy at > faqs.org: > http://www.faqs.org/faqs/C-faq/faq/ I think what the OP is saying is that if he adds the check for null, then his code works normally, including the file read operation, (ie, the pointer is not null), but if he removes the check, then he gets a segfault. Strange behavior like this is typically the result of memory corruption from earlier in the program. If you can post your entire code (preferably a minimal example that exhibits the behavior), then we can probably locate the problem. -Lewis -- 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: can't read sequential files
by Dave Korn
::
Rate this Message:
On 01 November 2007 15:15, Lewis Hyatt wrote:
>>> if (fp==NULL) >>> { >>> printf("error, NULL pointer!\n"); >>> return(1); >>> } > I think what the OP is saying is that if he adds the check for null, > then his code works normally, including the file read operation, (ie, > the pointer is not null), but if he removes the check, then he gets a > segfault. Please observe that the "check for NULL" also includes a return statement that bypasses the rest of the code .... including in particular the file read operation. 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: can't read sequential files
by Dave Korn
::
Rate this Message:
On 01 November 2007 15:58, Dave Korn wrote:
> On 01 November 2007 15:15, Lewis Hyatt wrote: > > >>>> if (fp==NULL) >>>> { >>>> printf("error, NULL pointer!\n"); >>>> return(1); >>>> } > >> I think what the OP is saying is that if he adds the check for null, >> then his code works normally, including the file read operation, (ie, >> the pointer is not null), but if he removes the check, then he gets a >> segfault. > > > Please observe that the "check for NULL" also includes a return statement > that bypasses the rest of the code .... including in particular the file > read operation. Hang on, I misread you, my eye skipped over the bit where you suggest that adding the check somehow makes the preceding fopen call succeed instead of fail. However I still don't think that's what the OP was saying, unless the subject line of this thread is terribly wrong, I think you just read a bit too much into OP's phrase "everything worked"; I think that just means "program ran to completion /without/ a segfault". 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: can't read sequential files
by Lewis Hyatt
::
Rate this Message:
> Hang on, I misread you, my eye skipped over the bit where you suggest that
> adding the check somehow makes the preceding fopen call succeed instead of > fail. However I still don't think that's what the OP was saying, unless the > subject line of this thread is terribly wrong, I think you just read a bit too > much into OP's phrase "everything worked"; I think that just means "program > ran to completion /without/ a segfault". It was more than that: >and then rebuilding the code, everything worked. But it's strange that >if I delete that code segment and never check whether the fp pointer is >NULL or not, I always get a segmentation fault. Can this be some kind >of an optimization problem? I don't know why it happens. Thank you for >the comments. Especially when he suggested it was a problem with the compiler optimizer, I am pretty sure that he means the code is working as expected with this check in there, and he doesn't understand why. (The subject line applies from back before he tried this.) But anyway you could be right too, I guess there's no point in guessing what he meant... -Lewis -- 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/ |
| Free embeddable forum powered by Nabble | Forum Help |
