netdev - [PATCH bpf-next 11/19] bpf: Disable migration in htab_map

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20250106081900.1665573-12-houtao@huaweicloud.com>
Date: Mon,  6 Jan 2025 16:18:52 +0800
From: Hou Tao <houtao@...weicloud.com>
To: bpf@...r.kernel.org,
	netdev@...r.kernel.org
Cc: Martin KaFai Lau <martin.lau@...ux.dev>,
	Alexei Starovoitov <alexei.starovoitov@...il.com>,
	Andrii Nakryiko <andrii@...nel.org>,
	Eduard Zingerman <eddyz87@...il.com>,
	Song Liu <song@...nel.org>,
	Hao Luo <haoluo@...gle.com>,
	Yonghong Song <yonghong.song@...ux.dev>,
	Daniel Borkmann <daniel@...earbox.net>,
	KP Singh <kpsingh@...nel.org>,
	Stanislav Fomichev <sdf@...ichev.me>,
	Jiri Olsa <jolsa@...nel.org>,
	John Fastabend <john.fastabend@...il.com>,
	houtao1@...wei.com,
	xukuohai@...wei.com
Subject: [PATCH bpf-next 11/19] bpf: Disable migration in htab_map_free()

From: Hou Tao <houtao1@...wei.com>

When freeing the hash map, the destroy procedure may invoke
bpf_obj_free_fields() to free the special fields in pre-allocated values
or dynamically-allocated values. Since these special fields may be
allocated from bpf memory allocator, migrate_{disable|enable} pairs are
necessary for the freeing of these objects.

To simplify reasoning about when migrate_disable() is needed for the
freeing of these dynamically-allocated objects, let the caller to
guarantee migration is disabled before invoking bpf_obj_free_fields().

For dynamically allocated values, delete_all_elements() already disables
migration before invoking bpf_obj_free_fields(). Therefore, the patch
moves migrate_{disable|enable} pair from delete_all_elements() to
htab_map_free() to handle all bpf_obj_free_fields() invocations. The
migrate_{disable|enable} pairs in the underlying implementation of
bpf_obj_free_fields() will be removed by the following patch.

Signed-off-by: Hou Tao <houtao1@...wei.com>
---
 kernel/bpf/hashtab.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 1db71d25836e..8bf1ad326e02 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -1502,10 +1502,9 @@ static void delete_all_elements(struct bpf_htab *htab)
 {
 	int i;
 
-	/* It's called from a worker thread, so disable migration here,
-	 * since bpf_mem_cache_free() relies on that.
+	/* It's called from a worker thread and migration has been disabled,
+	 * therefore, it is OK to invoke bpf_mem_cache_free() directly.
 	 */
-	migrate_disable();
 	for (i = 0; i < htab->n_buckets; i++) {
 		struct hlist_nulls_head *head = select_bucket(htab, i);
 		struct hlist_nulls_node *n;
@@ -1517,7 +1516,6 @@ static void delete_all_elements(struct bpf_htab *htab)
 		}
 		cond_resched();
 	}
-	migrate_enable();
 }
 
 static void htab_free_malloced_timers_and_wq(struct bpf_htab *htab)
@@ -1572,12 +1570,14 @@ static void htab_map_free(struct bpf_map *map)
 	 * underneath and is responsible for waiting for callbacks to finish
 	 * during bpf_mem_alloc_destroy().
 	 */
+	migrate_disable();
 	if (!htab_is_prealloc(htab)) {
 		delete_all_elements(htab);
 	} else {
 		htab_free_prealloced_fields(htab);
 		prealloc_destroy(htab);
 	}
+	migrate_enable();
 
 	bpf_map_free_elem_count(map);
 	free_percpu(htab->extra_elems);
-- 
2.29.2