From: John Hubbard <jhubb...@nvidia.com>
The page->dma_pinned_flags and _count fields require
lock protection. A lock at approximately the granularity
of the zone_lru_lock is called for, but adding to the
locking contention of zone_lru_lock is undesirable,
because that is a pre-existing hot spot. Fortunately,
these new dma_pinned_* fields can use an independent
lock, so this patch creates an entirely new lock, right
next to the zone_lru_lock.
Why "zone_gup_lock"?
Most of the naming refers to "DMA-pinned pages", but
"zone DMA lock" has other meanings already, so this is
called zone_gup_lock instead. The "dma pinning" is a result
of get_user_pages (gup) being called, so the name still
helps explain its use.
Signed-off-by: John Hubbard <jhubb...@nvidia.com>
---
include/linux/mmzone.h | 7 +++++++
mm/page_alloc.c | 1 +
2 files changed, 8 insertions(+)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 32699b2dc52a..5b4ceef82657 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -662,6 +662,8 @@ typedef struct pglist_data {
int kswapd_failures; /* Number of 'reclaimed == 0' runs */
+ spinlock_t pinned_dma_lock;
+
#ifdef CONFIG_COMPACTION
int kcompactd_max_order;
enum zone_type kcompactd_classzone_idx;
@@ -740,6 +742,11 @@ static inline spinlock_t *zone_lru_lock(struct zone *zone)
return &zone->zone_pgdat->lru_lock;
}
+static inline spinlock_t *zone_gup_lock(struct zone *zone)
+{
+ return &zone->zone_pgdat->pinned_dma_lock;
+}
+
static inline struct lruvec *node_lruvec(struct pglist_data *pgdat)
{
return &pgdat->lruvec;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 1521100f1e63..9c493442b57c 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -6211,6 +6211,7 @@ static void __paginginit free_area_init_core(struct
pglist_data *pgdat)
int nid = pgdat->node_id;
pgdat_resize_init(pgdat);
+ spin_lock_init(&pgdat->pinned_dma_lock);
#ifdef CONFIG_NUMA_BALANCING
spin_lock_init(&pgdat->numabalancing_migrate_lock);
pgdat->numabalancing_migrate_nr_pages = 0;
--
2.18.0
