Re: [Patch 17/23] mask v2 = [6/7] nodemask_t_ia64_changes

[Paul Jackson]

Several architectures have this large version of find_next_bit() code.

It may well make sense for the O(1) scheduler to be inlining this.

Perhaps these large versions should be renamed sched_find_next_bit(), as
is done for sched_find_first_bit(), and another, more space efficient
version provided under the name find_next_bit(), for use by others who
should prefer smaller code size.  The asm-generic version of this
smaller one might be something like [pseudo C]:

  int find_next_bit(addr, size, pos)
  {
	int i;
	for (i = pos; i < size; i++) {
		if (test_bit(i, addr))
			break;
	}
	return i;
  }

Arch's that have a reasonably tight version of find_next_bit() already
could use it for both sched_find_next_bit() and find_next_bit().

Just speculating ...

-- 
                          I won't rest till it's the best ...
                          Programmer, Linux Scalability
                          Paul Jackson <pj@xxxxxxx> 1.650.933.1373
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/