mm/slub: Reduce memory consumption in extreme scenarios

When kmalloc_node() is called without __GFP_THISNODE and the target node
lacks sufficient memory, SLUB allocates a folio from a different node
other than the requested node, instead of taking a partial slab from it.

However, since the allocated folio does not belong to the requested
node, on the following allocation it is deactivated and added to the
partial slab list of the node it belongs to.

This behavior can result in excessive memory usage when the requested
node has insufficient memory, as SLUB will repeatedly allocate folios
from other nodes without reusing the previously allocated ones.

To prevent memory wastage, when a preferred node is indicated (not
NUMA_NO_NODE) but without a prior __GFP_THISNODE constraint:

1) try to get a partial slab from target node only by having
   __GFP_THISNODE in pc.flags for get_partial()
2) if 1) failed, try to allocate a new slab from target node with
   GFP_NOWAIT | __GFP_THISNODE opportunistically.
3) if 2) failed, retry with original gfpflags which will allow
   get_partial() try partial lists of other nodes before potentially
   allocating new page from other nodes

Without a preferred node, or with __GFP_THISNODE constraint, the
behavior remains unchanged.

On qemu with 4 numa nodes and each numa has 1G memory. Write a test ko
to call kmalloc_node(196, GFP_KERNEL, 3) for (4 * 1024 + 4) * 1024 times.

cat /proc/slabinfo shows:
kmalloc-256       4200530 13519712    256   32    2 : tunables..

after this patch,
cat /proc/slabinfo shows:
kmalloc-256       4200558 4200768    256   32    2 : tunables..

Signed-off-by: Chen Jun <chenjun102@huawei.com>
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>

diff --git a/mm/slub.c b/mm/slub.c
index d35d918776250812a154c7e4a80b1ed3dfce1ca3..f152b5df8ab2e4a1f16aa74168c38f2991fa145a 100644 (file)
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -2699,7 +2699,7 @@ static struct slab *get_partial(struct kmem_cache *s, int node,
                searchnode = numa_mem_id();
 
        slab = get_partial_node(s, get_node(s, searchnode), pc);
-       if (slab || node != NUMA_NO_NODE)
+       if (slab || (node != NUMA_NO_NODE && (pc->flags & __GFP_THISNODE)))
                return slab;
 
        return get_any_partial(s, pc);
@@ -3375,6 +3375,7 @@ static void *___slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
        struct slab *slab;
        unsigned long flags;
        struct partial_context pc;
+       bool try_thisnode = true;
 
        stat(s, ALLOC_SLOWPATH);
 
@@ -3501,6 +3502,21 @@ new_slab:
 new_objects:
 
        pc.flags = gfpflags;
+       /*
+        * When a preferred node is indicated but no __GFP_THISNODE
+        *
+        * 1) try to get a partial slab from target node only by having
+        *    __GFP_THISNODE in pc.flags for get_partial()
+        * 2) if 1) failed, try to allocate a new slab from target node with
+        *    GPF_NOWAIT | __GFP_THISNODE opportunistically
+        * 3) if 2) failed, retry with original gfpflags which will allow
+        *    get_partial() try partial lists of other nodes before potentially
+        *    allocating new page from other nodes
+        */
+       if (unlikely(node != NUMA_NO_NODE && !(gfpflags & __GFP_THISNODE)
+                    && try_thisnode))
+               pc.flags = GFP_NOWAIT | __GFP_THISNODE;
+
        pc.orig_size = orig_size;
        slab = get_partial(s, node, &pc);
        if (slab) {
@@ -3522,10 +3538,15 @@ new_objects:
        }
 
        slub_put_cpu_ptr(s->cpu_slab);
-       slab = new_slab(s, gfpflags, node);
+       slab = new_slab(s, pc.flags, node);
        c = slub_get_cpu_ptr(s->cpu_slab);
 
        if (unlikely(!slab)) {
+               if (node != NUMA_NO_NODE && !(gfpflags & __GFP_THISNODE)
+                   && try_thisnode) {
+                       try_thisnode = false;
+                       goto new_objects;
+               }
                slab_out_of_memory(s, gfpflags, node);
                return NULL;
        }