Re: [PATCH v11 4/7] mm: introduce compaction and migration for ballooned pages
On Tue, Nov 20, 2012 at 03:33:24PM -0800, Andrew Morton wrote:
> On Fri, 9 Nov 2012 12:16:02 +
> Mel Gorman wrote:
>
> > On Wed, Nov 07, 2012 at 01:05:51AM -0200, Rafael Aquini wrote:
> > > Memory fragmentation introduced by ballooning might reduce significantly
> > > the number of 2MB contiguous memory blocks that can be used within a
> > > guest,
> > > thus imposing performance penalties associated with the reduced number of
> > > transparent huge pages that could be used by the guest workload.
> > >
> > > This patch introduces the helper functions as well as the necessary
> > > changes
> > > to teach compaction and migration bits how to cope with pages which are
> > > part of a guest memory balloon, in order to make them movable by memory
> > > compaction procedures.
> > >
> >
> > ...
> >
> > > --- a/mm/compaction.c
> > > +++ b/mm/compaction.c
> > > @@ -14,6 +14,7 @@
> > > #include
> > > #include
> > > #include
> > > +#include
> > > #include "internal.h"
> > >
> > > #if defined CONFIG_COMPACTION || defined CONFIG_CMA
> > > @@ -565,9 +566,24 @@ isolate_migratepages_range(struct zone *zone, struct
> > > compact_control *cc,
> > > goto next_pageblock;
> > > }
> > >
> > > - /* Check may be lockless but that's ok as we recheck later */
> > > - if (!PageLRU(page))
> > > + /*
> > > + * Check may be lockless but that's ok as we recheck later.
> > > + * It's possible to migrate LRU pages and balloon pages
> > > + * Skip any other type of page
> > > + */
> > > + if (!PageLRU(page)) {
> > > + if (unlikely(balloon_page_movable(page))) {
> >
> > Because it's lockless, it really seems that the barrier stuck down there
> > is unnecessary. At worst you get a temporarily incorrect answer that you
> > recheck later under page lock in balloon_page_isolate.
>
Sorry for the late reply.
> What happened with this?
>
This Mel's concern were addressed by the last submitted review (v12)
> Also: what barrier?
Mel was refering to these barriers, at balloon_compaction.h:
---8<---
+/*
+ * balloon_page_insert - insert a page into the balloon's page list and make
+ * the page->mapping assignment accordingly.
+ * @page: page to be assigned as a 'balloon page'
+ * @mapping : allocated special 'balloon_mapping'
+ * @head: balloon's device page list head
+ */
+static inline void balloon_page_insert(struct page *page,
+ struct address_space *mapping,
+ struct list_head *head)
+{
+ list_add(&page->lru, head);
+ /*
+* Make sure the page is already inserted on balloon's page list
+* before assigning its ->mapping.
+*/
+ smp_wmb();
+ page->mapping = mapping;
+}
+
+/*
+ * balloon_page_delete - clear the page->mapping and delete the page from
+ * balloon's page list accordingly.
+ * @page: page to be released from balloon's page list
+ */
+static inline void balloon_page_delete(struct page *page)
+{
+ page->mapping = NULL;
+ /*
+* Make sure page->mapping is cleared before we proceed with
+* balloon's page list deletion.
+*/
+ smp_wmb();
+ list_del(&page->lru);
+}
+
+/*
+ * __is_movable_balloon_page - helper to perform @page mapping->flags tests
+ */
+static inline bool __is_movable_balloon_page(struct page *page)
+{
+ /*
+* we might attempt to read ->mapping concurrently to other
+* threads trying to write to it.
+*/
+ struct address_space *mapping = ACCESS_ONCE(page->mapping);
+ smp_read_barrier_depends();
+ return mapping_balloon(mapping);
+}
---8<---
The last review got rid of them to stick with Mel's ACK.
Cheers!
--Rafael
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH v11 4/7] mm: introduce compaction and migration for ballooned pages
On Fri, 9 Nov 2012 12:16:02 +
Mel Gorman wrote:
> On Wed, Nov 07, 2012 at 01:05:51AM -0200, Rafael Aquini wrote:
> > Memory fragmentation introduced by ballooning might reduce significantly
> > the number of 2MB contiguous memory blocks that can be used within a guest,
> > thus imposing performance penalties associated with the reduced number of
> > transparent huge pages that could be used by the guest workload.
> >
> > This patch introduces the helper functions as well as the necessary changes
> > to teach compaction and migration bits how to cope with pages which are
> > part of a guest memory balloon, in order to make them movable by memory
> > compaction procedures.
> >
>
> ...
>
> > --- a/mm/compaction.c
> > +++ b/mm/compaction.c
> > @@ -14,6 +14,7 @@
> > #include
> > #include
> > #include
> > +#include
> > #include "internal.h"
> >
> > #if defined CONFIG_COMPACTION || defined CONFIG_CMA
> > @@ -565,9 +566,24 @@ isolate_migratepages_range(struct zone *zone, struct
> > compact_control *cc,
> > goto next_pageblock;
> > }
> >
> > - /* Check may be lockless but that's ok as we recheck later */
> > - if (!PageLRU(page))
> > + /*
> > +* Check may be lockless but that's ok as we recheck later.
> > +* It's possible to migrate LRU pages and balloon pages
> > +* Skip any other type of page
> > +*/
> > + if (!PageLRU(page)) {
> > + if (unlikely(balloon_page_movable(page))) {
>
> Because it's lockless, it really seems that the barrier stuck down there
> is unnecessary. At worst you get a temporarily incorrect answer that you
> recheck later under page lock in balloon_page_isolate.
What happened with this?
Also: what barrier?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH v11 4/7] mm: introduce compaction and migration for ballooned pages
On Wed, Nov 07, 2012 at 01:05:51AM -0200, Rafael Aquini wrote:
> Memory fragmentation introduced by ballooning might reduce significantly
> the number of 2MB contiguous memory blocks that can be used within a guest,
> thus imposing performance penalties associated with the reduced number of
> transparent huge pages that could be used by the guest workload.
>
> This patch introduces the helper functions as well as the necessary changes
> to teach compaction and migration bits how to cope with pages which are
> part of a guest memory balloon, in order to make them movable by memory
> compaction procedures.
>
> Signed-off-by: Rafael Aquini
> ---
> mm/compaction.c | 21 +++--
> mm/migrate.c| 36 ++--
> 2 files changed, 53 insertions(+), 4 deletions(-)
>
> diff --git a/mm/compaction.c b/mm/compaction.c
> index 9eef558..76abd84 100644
> --- a/mm/compaction.c
> +++ b/mm/compaction.c
> @@ -14,6 +14,7 @@
> #include
> #include
> #include
> +#include
> #include "internal.h"
>
> #if defined CONFIG_COMPACTION || defined CONFIG_CMA
> @@ -565,9 +566,24 @@ isolate_migratepages_range(struct zone *zone, struct
> compact_control *cc,
> goto next_pageblock;
> }
>
> - /* Check may be lockless but that's ok as we recheck later */
> - if (!PageLRU(page))
> + /*
> + * Check may be lockless but that's ok as we recheck later.
> + * It's possible to migrate LRU pages and balloon pages
> + * Skip any other type of page
> + */
> + if (!PageLRU(page)) {
> + if (unlikely(balloon_page_movable(page))) {
Because it's lockless, it really seems that the barrier stuck down there
is unnecessary. At worst you get a temporarily incorrect answer that you
recheck later under page lock in balloon_page_isolate.
> + if (locked && balloon_page_isolate(page)) {
> + /* Successfully isolated */
> + cc->finished_update_migrate = true;
> + list_add(&page->lru, migratelist);
> + cc->nr_migratepages++;
> + nr_isolated++;
> + goto check_compact_cluster;
> + }
> + }
> continue;
> + }
>
> /*
>* PageLRU is set. lru_lock normally excludes isolation
> @@ -621,6 +637,7 @@ isolate_migratepages_range(struct zone *zone, struct
> compact_control *cc,
> cc->nr_migratepages++;
> nr_isolated++;
>
> +check_compact_cluster:
> /* Avoid isolating too much */
> if (cc->nr_migratepages == COMPACT_CLUSTER_MAX) {
> ++low_pfn;
> diff --git a/mm/migrate.c b/mm/migrate.c
> index 98c7a89..87ffe54 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -35,6 +35,7 @@
> #include
> #include
> #include
> +#include
>
> #include
>
> @@ -79,7 +80,10 @@ void putback_lru_pages(struct list_head *l)
> list_del(&page->lru);
> dec_zone_page_state(page, NR_ISOLATED_ANON +
> page_is_file_cache(page));
> - putback_lru_page(page);
> + if (unlikely(balloon_page_movable(page)))
> + balloon_page_putback(page);
> + else
> + putback_lru_page(page);
> }
> }
>
> @@ -778,6 +782,18 @@ static int __unmap_and_move(struct page *page, struct
> page *newpage,
> }
> }
>
> + if (unlikely(balloon_page_movable(page))) {
> + /*
> + * A ballooned page does not need any special attention from
> + * physical to virtual reverse mapping procedures.
> + * Skip any attempt to unmap PTEs or to remap swap cache,
> + * in order to avoid burning cycles at rmap level, and perform
> + * the page migration right away (proteced by page lock).
> + */
> + rc = balloon_page_migrate(newpage, page, mode);
> + goto uncharge;
> + }
> +
> /*
>* Corner case handling:
>* 1. When a new swap-cache page is read into, it is added to the LRU
> @@ -814,7 +830,9 @@ skip_unmap:
> put_anon_vma(anon_vma);
>
> uncharge:
> - mem_cgroup_end_migration(mem, page, newpage, rc == MIGRATEPAGE_SUCCESS);
> + mem_cgroup_end_migration(mem, page, newpage,
> + (rc == MIGRATEPAGE_SUCCESS ||
> + rc == MIGRATEPAGE_BALLOON_SUCCESS));
> unlock:
> unlock_page(page);
> out:
> @@ -846,6 +864,20 @@ static int unmap_and_move(new_page_t get_new_page,
> unsigned long private,
> goto out;
>
>
