Re: [RFC] Avoiding fragmentation through different allocator

[Marcelo Tosatti]

On Wed, Jan 12, 2005 at 11:31:46PM -0800, William Lee Irwin III wrote:
> On Wed, Jan 12, 2005 at 09:09:24PM +0000, Mel Gorman wrote:
> > So... What the patch does. Allocations are divided up into three different
> > types of allocations;
> > UserReclaimable - These are userspace pages that are easily reclaimable. Right
> > 	now, I'm putting all allocations of GFP_USER and GFP_HIGHUSER as
> > 	well as disk-buffer pages into this category. These pages are trivially
> > 	reclaimed by writing the page out to swap or syncing with backing
> > 	storage
> > KernelReclaimable - These are pages allocated by the kernel that are easily
> > 	reclaimed. This is stuff like inode caches, dcache, buffer_heads etc.
> > 	These type of pages potentially could be reclaimed by dumping the
> > 	caches and reaping the slabs (drastic, but you get the idea). We could
> > 	also add pages into this category that are known to be only required
> > 	for a short time like buffers used with DMA
> > KernelNonReclaimable - These are pages that are allocated by the kernel that
> > 	are not trivially reclaimed. For example, the memory allocated for a
> > 	loaded module would be in this category. By default, allocations are
> > 	considered to be of this type
> 
> I'd expect to do better with kernel/user discrimination only, having
> address-ordering biases in opposite directions for each case.

What you mean with "address-ordering biases in opposite directions for each case" ? 

You mean to have each case allocate from the top and bottom of the free list, respectively,
and in opposite address direction ? What you gain from that?

And what that means during a long period of VM stress ?

-
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/