mike3 wrote:
> Hi.
>
> I was using mmap() in Solaris 10 Unix and I've been having some
> trouble. When I go and
> mmap() a file of a given length, then pass that to a routine in the
> program with that
> exact same data length (i.e. the length of the file and mmaped region
> are equal), it fails
> with a "Bus error", "object specific error". Yet if I add as little as
> 1 byte to the end of the
> file, it works OK, no more bus error.
>
Posting the code would help.

--
Ian Collins.

mike3 writes:

> I was using mmap() in Solaris 10 Unix and I've been having some
> trouble. When I go and mmap() a file of a given length, then pass
> that to a routine in the program with that exact same data length
> (i.e. the length of the file and mmaped region are equal), it fails
> with a "Bus error", "object specific error".

Generally that means that you've mmap()ed more pages then there
are in the file. From "man mmap":

The mmap() function allows [pa, pa + len) to extend beyond
the end of the object both at the time of the mmap() and
while the mapping persists, such as when the file is created
prior to the mmap() call and has no contents, or when the
file is truncated. Any reference to addresses beyond the end
of the object, however, will result in the delivery of a
SIGBUS or SIGSEGV signal. The mmap() function cannot be used
to implicitly extend the length of files.


> Yet if I add as little as 1 byte to the end of the
> file, it works OK, no more bus error.

Sounds like you have a file with size == N*getpagesize(), and you are
miscalculating its size by 1 page.

Wrong code to round size up to whole page:

int pagesize = getpagesize();
off_t sz = pagesize * (st.st_size / pagesize + 1);
// sz is correct, except when st.st_size % pagesize == 0

Correct:

off_t sz = (st.st_size + pagesize - 1) & ~(pagesize - 1)

> P.S. Data length is 16,777,216 bytes.

Confirms my theory: 16777216 == 0x1000000

Cheers,
--
In order to understand recursion you must first understand recursion.
Remove /-nsp/ for email.