This is the mail archive of the cygwin mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: second call to mmap() results in error


On Jan 30 07:25, Steven Bardwell wrote:
> > > On 29/01/2014 19:12, Corinna Vinschen wrote:
> > > >On Jan 29 09:00, Steven Bardwell wrote:
> > > >>My application needs several areas of shared memory, and I am getting
> > an
> > > >>error ("No such device") on the second call to mmap(). The first call
> > works
> > > >>fine.
> > 
> > Sorry guys, but it still works fine for me.  I tried your testcase on W7
> > 32, W7 64 in 32 and 64 bit, and on Windows 8.1 64 in 32 and 64 bit.  I
> > tried it with Cygwin 1.7.27 and with the latest snapshot.  I'm always
> > getting the output "Shared memory initialized" and no error at all.
> > 
> > 
> > Any chance one of you guys could debug this further, by stepping through
> > the Cygwin mmap64 function, preferredly using the latest snapshot or,
> > a self-built Cygwin DLL from?
> > 
> > 
> > Corinna
> 
> I reinstalled Cygwin, rebooted and the error persisted. Running 'gdb' and
> stepping through the program showed that the call to mmap() fails for /block1
> also -- it is returning an invalid address. This simplification of the program
> shows that error on my machine ('Bus error (core dumped)' ) occurs
> when it tries to do the memcpy() to the mapped address.

Try this:

> #include <stdio.h>
> #include <stdlib.h>

  #include <unistd.h>

> #include <string.h>
> #include <sys/types.h>
> #include <sys/errno.h>
> #include <fcntl.h>
> #include <sys/mman.h>
> 
> int main()
> {
>   int shm_fd1;
>   char *mmap1;
>   shm_fd1 = shm_open("/block1", O_CREAT | O_RDWR, 0666);
>   if (shm_fd1 == -1) {
>     fprintf(stderr, "Couldn't get fd for block1 (%s)\n", strerror(errno));
>     exit(1);
>   }
>   ftruncate(shm_fd1, 524304);
>   mmap1 = mmap(NULL, 524304, PROT_WRITE | PROT_READ, MAP_SHARED, shm_fd1, 0);
>   if (mmap1 == (char *)-1) {
>     fprintf(stderr, "Couldn't map memory for /block1 (%s)\n", strerror(errno));
>     exit(1);
>   }
>   memcpy(mmap1, "ABCDEF\0", 7);
>   fprintf(stdout, mmap1);
> 
>   fprintf(stdout, "Shared memory initialized\n");
>   exit(0);
> }

The reason is that ftruncate is defined with the second argument being
off_t, which is 8 byte.  524304 is an int (4 byte) only, though.  Since
ftruncate is declared in unistd.h, but you didn't include unistd.h, the
2nd parameter to ftruncate is auto-propagated to int, which results in
an invalid new file length, and which makes ftruncate fail.  Since you
missed to check ftruncate's return value... you get the idea.


Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Maintainer                 cygwin AT cygwin DOT com
Red Hat

Attachment: pgp6fXD_epSsr.pgp
Description: PGP signature


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]