>> I've done a little more looking at the 'rusting' of the memory on
>> my machine and I've reached a view:
> Noted. This is the memory fragmentation problem for which we're
> trying to decide which means of attack can be used for 2.2. Just
> keep an eye out for patches addressing fragmentation (don't ask for
> an exact time frame -- my current play is nowhere near ready for
> prime time yet).
One idea would be to have the swap manager logic set so that when it
reads some swapped-out pages back in, it tries to place them in one or
more of the following places, as appropriate...
1. As near to the rest of the same application as possible.
2. Where there's a free block that's exactly the right size.
3. At one end of the smallest free block available that's large
enough to hold it.
Also, where it can satisfy two or more of those requirements, it does
so in preference to only satisfying one...
This should have the effect of reducing the fragmentation of the free
blocks with very little overhead...
Just my $0.02 worth...and probably already implemented, knowing my
luck !!!
Best wishes from Riley.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu