Re: [RFC][DATA] re "ongoing vm suckage"

[Marcelo Tosatti <marcelo@conectiva.com.br>]

On Tue, 7 Aug 2001, Linus Torvalds wrote:

> On Tue, 7 Aug 2001, Linus Torvalds wrote:
> >
> > Sorry, I should have warned people: pre5 is a test-release that was
> > intended solely for Leonard Zubkoff who has been helping with trying to
> > debug a FS livelock condition.
> 
> In case people are interested, the livelock _seems_ to be rather simple:
> when we run out of immediately available memory in "refill_freelist()", we
> seem to do all the wrong things.
> 
> This condition doesn't happen under normal load, because under normal load
> most allocators are of the regular "GFP_USER" kind. But it looks like
> under certain loads you get into the situation that the system isn't doing
> much else than just doing alloc_pages(GFP_NOFS), and that one seems to be
> giving up really easily.
> 
> Apparently so easily, in fact, that kswapd doesn't even bother to try to
> make things better. I _know_ that it doesn't show up as
> "inactive_shortage()", because the machine has tons of inactive_dirty
> pages (buffers), and I suspect that "free_shortage()" also decides that we
> have enough pages.
> 
> So I _think_ that what happens is:
>  - alloc_pages() itself isn't making any progress, because it's called
>    with GFP_NOFS and thus cannot touch a lot of the pages.
>  - we wake up kswapd to try to help, but kswapd doesn't do anything
>    because it thinks things are fine.

If the reason for this problem is really what you're thinking, the
following together with the total_free_shortage() patch should fix the
problem.

(please apply this on the stock tree. Even if it does not fix the problem
we're seeing now, it is correct)

--- mm/page_alloc.c.origf       Wed Aug  8 00:45:51 2001
+++ mm/page_alloc.c     Wed Aug  8 00:46:19 2001
@@ -453,7 +453,7 @@
                if (gfp_mask & __GFP_WAIT) {
                        if (!order || total_free_shortage()) {
                                int progress = try_to_free_pages(gfp_mask);
-                               if (progress || (gfp_mask & __GFP_FS))
+                               if (progress || (gfp_mask & __GFP_IO))
                                        goto try_again;
                                /*
                                 * Fail in case no progress was made and the


--
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/