Puzzled by __vm_enough_memory with OVERCOMMIT_NEVER

[VomLehn <dvomlehn@cisco.com>]

The function __vm_enough_memory in mm/mmap.c has a piece of code for
handling the case of disabled overcommit that has puzzled me for a while
and looks like it may be causing a problem:

	/* Don't let a single process grow too big:
	   leave 3% of the size of this process for other processes */
	if (mm)
		allowed -= mm->total_vm / 32;

At this point, it seems like total_vm does not yet include the pages
we are trying to add, so this is limiting a single process to no more than
97% of its *old* size rather than its new size. So, this seems to make more
sense:

	if (mm)
		allowed -= (mm->total_vm + pages)/ 32;

Even then, it seems like the real way to do this would be simply to lop off
3% of the total available virtual memory, and use:

	allowed -= allowed / 32;

Or, perhaps I'm missing something.
--
David VomLehn

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>