On Wed, 25 Jun 2003, Ronald Landheer-Cieslak wrote:
> On Wed, 25 Jun 2003, Elfyn McBratney wrote:
>
> > On Wed, 25 Jun 2003, Alex Vinokur wrote:
> >
> > > ===========================================
> > > Windows 2000
> > > CYGWIN_NT-5.0 1.3.22(0.78/3/2)
> > > GNU gcc version 3.2 20020927 (prerelease)
> > > ===========================================
> > >
> > > Here is some function.
> > >
> > > --------------------------------------
> > > void read_file (char* filename_i)
> > > {
> > > int fd = open(filename_i, O_RDONLY);
> > > assert (fd > 2);
> > >
> > > off_t sz = lseek(fd, 0, SEEK_END);
> > > char* ptr = (char*)mmap(0, sz, PROT_READ, 0, fd, 0);
> > >
> > > assert (ptr != MAP_FAILED); // Here assertion failed
> > > if (ptr != MAP_FAILED)
> > > {
> > > string str(ptr, ptr+sz);
> > > munmap(ptr, sz);
> > > }
> > >
> > > close(fd);
> > > }
> > > --------------------------------------
> > >
> > > Assertion "ptr != MAP_FAILED)" failed.
> > > What might cause that?
> >
> > This is just a stab in the dark, of course, but surely `ptr != MAP_FAILED'
> > would indicate that the mmap did not fail? Assertions (assert()) are based on
> > true or false, so the above assert is false in that `ptr != MAP_FAILED'.
> >
> > Elfyn
> >
> >
> Ehm..
>
> If ptr != MAP_FAILED is not true, that means ptr == MAP_FAILED.
> assert(ptr != MAP_FAILED) thus fails if mmap fails..
>
> unless I'm missing something..
But what's the point in 'assert (ptr != NULL)'? Surely the best way to test for
failure is the other way round (s/!=/==/), if ptr == MAP_FAILED assert, or
continue...No?
Elfyn
--
--
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/
