From: Rebecca Schultz Zavin <[email protected]>

When allocations larger than order 4 are made, use _GFP_NORETRY
and __GFP_NO_KSWAPD so kswapd doesn't get kicked off to reclaim
these larger chunks.  For smaller allocaitons, these are
unnecessary, as the system should be able to reclaim these.

Signed-off-by: Rebecca Schultz Zavin <[email protected]>
[jstultz: modified patch to apply to staging directory]
Signed-off-by: John Stultz <[email protected]>
---
 drivers/staging/android/ion/ion_system_heap.c | 25 ++++++++++++++++++-------
 1 file changed, 18 insertions(+), 7 deletions(-)

diff --git a/drivers/staging/android/ion/ion_system_heap.c 
b/drivers/staging/android/ion/ion_system_heap.c
index 35fbfff..b99fa0e 100644
--- a/drivers/staging/android/ion/ion_system_heap.c
+++ b/drivers/staging/android/ion/ion_system_heap.c
@@ -26,6 +26,11 @@
 #include "ion.h"
 #include "ion_priv.h"
 
+static unsigned int high_order_gfp_flags = (GFP_HIGHUSER | __GFP_ZERO |
+                                           __GFP_NOWARN | __GFP_NORETRY |
+                                           __GFP_NO_KSWAPD);
+static unsigned int low_order_gfp_flags  = (GFP_HIGHUSER | __GFP_ZERO |
+                                        __GFP_NOWARN);
 static const unsigned int orders[] = {8, 4, 0};
 static const int num_orders = ARRAY_SIZE(orders);
 static int order_to_index(unsigned int order)
@@ -63,11 +68,15 @@ static struct page *alloc_buffer_page(struct 
ion_system_heap *heap,
        struct ion_page_pool *pool = heap->pools[order_to_index(order)];
        struct page *page;
 
-       if (!cached)
+       if (!cached) {
                page = ion_page_pool_alloc(pool);
-       else
-               page = alloc_pages(GFP_HIGHUSER | __GFP_ZERO |
-                                  __GFP_NOWARN | __GFP_NORETRY, order);
+       } else {
+               gfp_t gfp_flags = low_order_gfp_flags;
+
+               if (order > 4)
+                       gfp_flags = high_order_gfp_flags;
+               page = alloc_pages(gfp_flags, order);
+       }
        if (!page)
                return 0;
        if (split_pages)
@@ -342,9 +351,11 @@ struct ion_heap *ion_system_heap_create(struct 
ion_platform_heap *unused)
                goto err_alloc_pools;
        for (i = 0; i < num_orders; i++) {
                struct ion_page_pool *pool;
-               pool = ion_page_pool_create(GFP_HIGHUSER | __GFP_ZERO |
-                                          __GFP_NOWARN | __GFP_NORETRY,
-                                          orders[i]);
+               gfp_t gfp_flags = low_order_gfp_flags;
+
+               if (orders[i] > 4)
+                       gfp_flags = high_order_gfp_flags;
+               pool = ion_page_pool_create(gfp_flags, orders[i]);
                if (!pool)
                        goto err_create_pool;
                heap->pools[i] = pool;
-- 
1.8.3.2

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

Reply via email to