Date:   Sun, 28 Jun 2020 19:43:45 +1200
From:   Barry Song <song.bao.hua@...ilicon.com>
To:     <akpm@...ux-foundation.org>
CC:     <linux-mm@...ck.org>, <linux-kernel@...r.kernel.org>,
        <linuxarm@...wei.com>, Barry Song <song.bao.hua@...ilicon.com>,
        "Jonathan Cameron" <Jonathan.Cameron@...wei.com>,
        Aslan Bakirov <aslan@...com>, "Roman Gushchin" <guro@...com>,
        Michal Hocko <mhocko@...nel.org>,
        Andreas Schaufler <andreas.schaufler@....de>,
        Mike Kravetz <mike.kravetz@...cle.com>,
        "Rik van Riel" <riel@...riel.com>, Joonsoo Kim <js1304@...il.com>,
        Robin Murphy <robin.murphy@....com>
Subject: [PATCH] mm/cma.c: use exact_nid true to fix possible per-numa cma leak

Calling cma_declare_contiguous_nid() with false exact_nid for per-numa
reservation can easily cause cma leak and various confusion.
For example, mm/hugetlb.c is trying to reserve per-numa cma for gigantic
pages. But it can easily leak cma and make users confused when system has
memoryless nodes.

In case the system has 4 numa nodes, and only numa node0 has memory.
if we set hugetlb_cma=4G in bootargs, mm/hugetlb.c will get 4 cma areas
for 4 different numa nodes. since exact_nid=false in current code, all
4 numa nodes will get cma successfully from node0, but hugetlb_cma[1 to 3]
will never be available to hugepage will only allocate memory from
hugetlb_cma[0].

In case the system has 4 numa nodes, both numa node0&2 has memory, other
nodes have no memory.
if we set hugetlb_cma=4G in bootargs, mm/hugetlb.c will get 4 cma areas
for 4 different numa nodes. since exact_nid=false in current code, all
4 numa nodes will get cma successfully from node0 or 2, but hugetlb_cma[1]
and [3] will never be available to hugepage as mm/hugetlb.c will only
allocate memory from hugetlb_cma[0] and hugetlb_cma[2].
This causes permanent leak of the cma areas which are supposed to be
used by memoryless node.

Of cource we can workaround the issue by letting mm/hugetlb.c scan all
cma areas in alloc_gigantic_page() even node_mask includes node0 only.
that means when node_mask includes node0 only, we can get page from
hugetlb_cma[1] to hugetlb_cma[3]. But this will cause kernel crash in
free_gigantic_page() while it wants to free page by:
cma_release(hugetlb_cma[page_to_nid(page)], page, 1 << order)

On the other hand, exact_nid=false won't consider numa distance, it
might be not that useful to leverage cma areas on remote nodes.
I feel it is much simpler to make exact_nid true to make everything
clear. After that, memoryless nodes won't be able to reserve per-numa
CMA from other nodes which have memory.

Fixes: cf11e85fc08c ("mm: hugetlb: optionally allocate gigantic hugepages using cma")
Cc: Jonathan Cameron <Jonathan.Cameron@...wei.com>
Cc: Aslan Bakirov <aslan@...com>
Cc: Roman Gushchin <guro@...com>
Cc: Andrew Morton <akpm@...ux-foundation.org>
Cc: Michal Hocko <mhocko@...nel.org>
Cc: Andreas Schaufler <andreas.schaufler@....de>
Cc: Mike Kravetz <mike.kravetz@...cle.com>
Cc: Rik van Riel <riel@...riel.com>
Cc: Joonsoo Kim <js1304@...il.com>
Cc: Robin Murphy <robin.murphy@....com>
Signed-off-by: Barry Song <song.bao.hua@...ilicon.com>
---
 mm/cma.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/cma.c b/mm/cma.c
index b24151fa2101..f472f398026f 100644
--- a/mm/cma.c
+++ b/mm/cma.c
@@ -338,13 +338,13 @@ int __init cma_declare_contiguous_nid(phys_addr_t base,
 		 */
 		if (base < highmem_start && limit > highmem_start) {
 			addr = memblock_alloc_range_nid(size, alignment,
-					highmem_start, limit, nid, false);
+					highmem_start, limit, nid, true);
 			limit = highmem_start;
 		}
 
 		if (!addr) {
 			addr = memblock_alloc_range_nid(size, alignment, base,
-					limit, nid, false);
+					limit, nid, true);
 			if (!addr) {
 				ret = -ENOMEM;
 				goto err;
-- 
2.27.0

Powered by blists - more mailing lists