Commit 48b4800a1c6a ("zsmalloc: page migration support") added support for migrating zsmalloc pages using the movable_operations migration framework. However, the commit did not take into account that zsmalloc supports migration only when CONFIG_COMPACTION is enabled. Tracing shows that zsmalloc was still passing the __GFP_MOVABLE flag even when compaction is not supported.
This can result in unmovable pages being allocated from movable page blocks (even without stealing page blocks), ZONE_MOVABLE and CMA area.
Clear the __GFP_MOVABLE flag when !IS_ENABLED(CONFIG_COMPACTION).
Cc: stable@vger.kernel.org Fixes: 48b4800a1c6a ("zsmalloc: page migration support") Signed-off-by: Harry Yoo harry.yoo@oracle.com --- mm/zsmalloc.c | 3 +++ 1 file changed, 3 insertions(+)
diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c index 999b513c7fdf..f3e2215f95eb 100644 --- a/mm/zsmalloc.c +++ b/mm/zsmalloc.c @@ -1043,6 +1043,9 @@ static struct zspage *alloc_zspage(struct zs_pool *pool, if (!zspage) return NULL;
+ if (!IS_ENABLED(CONFIG_COMPACTION)) + gfp &= ~__GFP_MOVABLE; + zspage->magic = ZSPAGE_MAGIC; zspage->pool = pool; zspage->class = class->index;
On Fri, Jul 04, 2025 at 07:30:53PM +0900, Harry Yoo wrote:
Commit 48b4800a1c6a ("zsmalloc: page migration support") added support for migrating zsmalloc pages using the movable_operations migration framework. However, the commit did not take into account that zsmalloc supports migration only when CONFIG_COMPACTION is enabled. Tracing shows that zsmalloc was still passing the __GFP_MOVABLE flag even when compaction is not supported.
This can result in unmovable pages being allocated from movable page blocks (even without stealing page blocks), ZONE_MOVABLE and CMA area.
Clear the __GFP_MOVABLE flag when !IS_ENABLED(CONFIG_COMPACTION).
Cc: stable@vger.kernel.org Fixes: 48b4800a1c6a ("zsmalloc: page migration support") Signed-off-by: Harry Yoo harry.yoo@oracle.com
Possible user visible effects: - Some ZONE_MOVABLE memory can be not actually movable - CMA allocation can fail because of this - Increased memory fragmentation due to ignoring the page mobility grouping feature
I'm not really sure who uses kernels without compaction support, though :(
mm/zsmalloc.c | 3 +++ 1 file changed, 3 insertions(+)
diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c index 999b513c7fdf..f3e2215f95eb 100644 --- a/mm/zsmalloc.c +++ b/mm/zsmalloc.c @@ -1043,6 +1043,9 @@ static struct zspage *alloc_zspage(struct zs_pool *pool, if (!zspage) return NULL;
- if (!IS_ENABLED(CONFIG_COMPACTION))
gfp &= ~__GFP_MOVABLE;
- zspage->magic = ZSPAGE_MAGIC; zspage->pool = pool; zspage->class = class->index;
-- 2.43.0
On 04.07.25 12:30, Harry Yoo wrote:
Commit 48b4800a1c6a ("zsmalloc: page migration support") added support for migrating zsmalloc pages using the movable_operations migration framework. However, the commit did not take into account that zsmalloc supports migration only when CONFIG_COMPACTION is enabled. Tracing shows that zsmalloc was still passing the __GFP_MOVABLE flag even when compaction is not supported.
This can result in unmovable pages being allocated from movable page blocks (even without stealing page blocks), ZONE_MOVABLE and CMA area.
Clear the __GFP_MOVABLE flag when !IS_ENABLED(CONFIG_COMPACTION).
Cc: stable@vger.kernel.org Fixes: 48b4800a1c6a ("zsmalloc: page migration support") Signed-off-by: Harry Yoo harry.yoo@oracle.com
mm/zsmalloc.c | 3 +++ 1 file changed, 3 insertions(+)
diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c index 999b513c7fdf..f3e2215f95eb 100644 --- a/mm/zsmalloc.c +++ b/mm/zsmalloc.c @@ -1043,6 +1043,9 @@ static struct zspage *alloc_zspage(struct zs_pool *pool, if (!zspage) return NULL;
- if (!IS_ENABLED(CONFIG_COMPACTION))
gfp &= ~__GFP_MOVABLE;
- zspage->magic = ZSPAGE_MAGIC; zspage->pool = pool; zspage->class = class->index;
IIUC, the real problem was exposed once zsmalloc allocation users started passing __GFP_MOVABLE.
Probably zpool_malloc_support_movable() was used for determining that at some point.
Acked-by: David Hildenbrand david@redhat.com
linux-stable-mirror@lists.linaro.org