On Tue, Jan 19, 2010 at 03:18:11PM +0200, Pekka Enberg wrote:

On Tue, Jan 19, 2010 at 2:48 PM, Minchan Kim [off-list ref] wrote:

quoted

Gleb. How about using MADV_SEQUENTIAL on guest memory?
It makes that pages of guest are moved into inactive reclaim list more
fast. It means it is likely to swap out faster than other pages if it
isn't hit during inactive list.

Yeah, something like that but we don't want the readahead. OTOH, it's
not clear what Gleb's real problem is. Are the guest address spaces
anonymous or file backed?

Anonymous.

                          Which parts of the emulator are swapped out
that are causing the problem?

I don't want anything that can be used during guest runtime to be
swapped out. And I run 2G guest in 512M container, so eventually
everything is swapped out :)

                               Maybe it's a VM balancing issue that
mlock papers over?

There is no problem. I do measurements on how host swapping affects
guest and I don't want qemu code to be swapped out.

--
			Gleb.

--
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>