On Mon, Jun 27, 2016 at 11:30:52AM +0200, Vlastimil Babka wrote:

On 05/26/2016 08:22 AM, js1304@gmail.com wrote:

quoted

From: Joonsoo Kim <redacted>

Now, all reserved pages for CMA region are belong to the ZONE_CMA
and it only serves for GFP_HIGHUSER_MOVABLE. Therefore, we don't need to
consider ALLOC_CMA at all.

Signed-off-by: Joonsoo Kim <redacted>
---
mm/internal.h   |  3 +--
mm/page_alloc.c | 27 +++------------------------
2 files changed, 4 insertions(+), 26 deletions(-)

[...]

quoted

@@ -2833,10 +2827,8 @@ bool __zone_watermark_ok(struct zone *z, unsigned int order, unsigned long mark,

		}

#ifdef CONFIG_CMA
-		if ((alloc_flags & ALLOC_CMA) &&
-		    !list_empty(&area->free_list[MIGRATE_CMA])) {
+		if (!list_empty(&area->free_list[MIGRATE_CMA]))
			return true;
-		}
#endif

Nitpick: it would be more logical to remove the whole block in this
patch, as removing ALLOC_CMA means it's effectively false? Also less
churn.

No, all freepages on ZONE_CMA is attached on area->free_list[MIGRATE_CMA].
We need to check whether there is a freepage on it or not to pass watermark
check for high-order allocation.

Thanks.

--
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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>