>>>>> "jesse" == Jesse Pollard <pollard@tomcat.admin.navo.hpc.mil> writes:
jesse> If I follow this right:
jesse> The first process fires off 4 processes.
jesse> Each of the second level processes fires off 3 processes
jesse> Each of the third level process fires off 2 processes
jesse> Each of the forth level processes fires of 1 process
jesse> total of 1 +(4 + 4 * (3 + 3 * (2 + 2 * (1)))) = 65 processes
jesse> Each process allocates 4MB array (COW) in the array arr, which is also
jesse> on the stack. (Each pointer occupies 4 bytes)
jesse> Each pointer points to an allocated 1K (unused).
jesse> 65 processes * 4MB => 260MB (minimum). There is some additional since the
jesse> malloc is allocating chunks of less than a page size, so I believe that
jesse> the 1 million pointers will point to 1M * 1KB => 1024MB data. One process
jesse> should then have 1025MB of data
Yes, I know that we are allocating a *lot* of memory, more than we
have. I don't expect to succeed all the programs, but I don't expect
to freeze the system. Note that:
1 - We don't touch any of the memory that we allocate with
malloc, I will expect that memory to not be reserved. It
should reserve the memory for the pointers/internal
bookeping, ... but not the malloced memory.
2 - It will die, that is ok with me, but it gets killed, and
other processes in the system are killed, or the system
hangs, I hope this process to be the only one being killed,
in my machine, this are the biggers programs, the programs
that are allocating more space, ....
And here, at least the system becomes completely unresponsive and then
begin to kill processes.
Later, Juan.
-- In theory, practice and theory are the same, but in practice they are different -- Larry McVoy- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue May 23 2000 - 21:00:25 EST