lkml.org 
[lkml]   [2005]   [Nov]   [30]   [last100]   RSS Feed
Views: [more markup]  [less markup]  [headers]  [forward] 
 
Messages in this thread
/
DateWed, 30 Nov 2005 17:34:47 +0900
FromKAMEZAWA Hiroyuki <>
SubjectRe: [Lhms-devel] [PATCH 4/7] Direct Migration V5: migrate_pages() extension
Hi,

Christoph Lameter wrote:
> +int migrate_page_remove_references(struct page *newpage, struct page *page, int nr_refs)
> +{
> +	write_lock_irq(&mapping->tree_lock);
> +
> +	radix_pointer = (struct page **)radix_tree_lookup_slot(
> +						&mapping->page_tree,
> +						page_index(page));
> +
> +	if (!page->mapping ||
> +	    page_count(page) != nr_refs ||
> +	    *radix_pointer != page) {
> +		write_unlock_irq(&mapping->tree_lock);
> +		return 1;
> +	}

I'm testing memory hot removing patch based on your patch.

I found a problem around the shmem,
but I'm not sure whether it can be problem on migration or not.

Problem is:
1. a page of shmem(tmpfs)'s generic file is in page-cache. assume page is diry.
2. When it passed to migrate_page(), it reaches pageout() in the middle of migrate_page().
3. pageout calls shmem_writepage(), and the page turns to be swap-cache page.
    At this point, page->mapping becomes NULL (see move_to_swapcache())
4. pageout retunrs PAGE_SUCCESS.
5. Finaly, migrate_page() goes to redo.
6. retry
7. Because spwapper_space's  a_ops->migratepage is not NULL,
    "Avoid write back hook" in patch 7/7 is used.
+		if (mapping->a_ops->migratepage) {
+			rc = mapping->a_ops->migratepage(newpage, page);
+			goto unlock_both;
+                }
    a_ops->migrate_page points to migrate_page() in mm/vmscan.c
8. migrate_page() try to replace radix tree entry in swapper_space.
9. Becasue page->mapping is NULL(becasue of 3), migrate_page_remove_references() fails.
I avoid above situation by following code in migrate_page_remove_references() now.
But I'm not sure whether this is sane fix or not.
 > +	if ((!PageSwapCache(page) && !page->mapping) ||
 > +	    page_count(page) != nr_refs ||
 > +	    *radix_pointer != page) {
 > +		write_unlock_irq(&mapping->tree_lock);
 > +		return 1;
 > +	}



-- Kame

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-11-30 08:40    [from the cache]
©2003-2008