Message-ID: <20241030222819.1800667-1-namhyung@kernel.org>
Date: Wed, 30 Oct 2024 15:28:18 -0700
From: Namhyung Kim <namhyung@...nel.org>
To: Alexei Starovoitov <ast@...nel.org>,
	Daniel Borkmann <daniel@...earbox.net>,
	Andrii Nakryiko <andrii@...nel.org>
Cc: Martin KaFai Lau <martin.lau@...ux.dev>,
	Eduard Zingerman <eddyz87@...il.com>,
	Song Liu <song@...nel.org>,
	Yonghong Song <yonghong.song@...ux.dev>,
	John Fastabend <john.fastabend@...il.com>,
	KP Singh <kpsingh@...nel.org>,
	Stanislav Fomichev <sdf@...ichev.me>,
	Hao Luo <haoluo@...gle.com>,
	Jiri Olsa <jolsa@...nel.org>,
	LKML <linux-kernel@...r.kernel.org>,
	bpf@...r.kernel.org,
	Andrew Morton <akpm@...ux-foundation.org>,
	Christoph Lameter <cl@...ux.com>,
	Pekka Enberg <penberg@...nel.org>,
	David Rientjes <rientjes@...gle.com>,
	Joonsoo Kim <iamjoonsoo.kim@....com>,
	Vlastimil Babka <vbabka@...e.cz>,
	Roman Gushchin <roman.gushchin@...ux.dev>,
	Hyeonggon Yoo <42.hyeyoo@...il.com>,
	linux-mm@...ck.org,
	Arnaldo Carvalho de Melo <acme@...nel.org>,
	Kees Cook <kees@...nel.org>
Subject: [PATCH bpf-next v3 1/2] bpf: Add open coded version of kmem_cache iterator

Add a new open coded iterator for kmem_cache which can be called from a
BPF program like below.  It doesn't take any argument and traverses all
kmem_cache entries.

  struct kmem_cache *pos;

  bpf_for_each(kmem_cache, pos) {
      ...
  }

As it needs to grab slab_mutex, it should be called from sleepable BPF
programs only.

Also update the existing iterator code to use the open coded version
internally as suggested by Andrii.

Signed-off-by: Namhyung Kim <namhyung@...nel.org>
---
v2)
 * prevent restart after the last element  (Martin)
 * update existing code to use the open coded version (Andrii)

 kernel/bpf/helpers.c         |   3 +
 kernel/bpf/kmem_cache_iter.c | 151 +++++++++++++++++++++++++----------
 2 files changed, 110 insertions(+), 44 deletions(-)

diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index 2e82f8d3a76fb9ca..395221e53832e10e 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -3112,6 +3112,9 @@ BTF_ID_FLAGS(func, bpf_iter_bits_next, KF_ITER_NEXT | KF_RET_NULL)
 BTF_ID_FLAGS(func, bpf_iter_bits_destroy, KF_ITER_DESTROY)
 BTF_ID_FLAGS(func, bpf_copy_from_user_str, KF_SLEEPABLE)
 BTF_ID_FLAGS(func, bpf_get_kmem_cache)
+BTF_ID_FLAGS(func, bpf_iter_kmem_cache_new, KF_ITER_NEW | KF_SLEEPABLE)
+BTF_ID_FLAGS(func, bpf_iter_kmem_cache_next, KF_ITER_NEXT | KF_RET_NULL | KF_SLEEPABLE)
+BTF_ID_FLAGS(func, bpf_iter_kmem_cache_destroy, KF_ITER_DESTROY | KF_SLEEPABLE)
 BTF_KFUNCS_END(common_btf_ids)
 
 static const struct btf_kfunc_id_set common_kfunc_set = {
diff --git a/kernel/bpf/kmem_cache_iter.c b/kernel/bpf/kmem_cache_iter.c
index ebc101d7da51b57c..3ae2158d767f4526 100644
--- a/kernel/bpf/kmem_cache_iter.c
+++ b/kernel/bpf/kmem_cache_iter.c
@@ -8,16 +8,116 @@
 
 #include "../../mm/slab.h" /* kmem_cache, slab_caches and slab_mutex */
 
+/* open-coded version */
+struct bpf_iter_kmem_cache {
+	__u64 __opaque[1];
+} __attribute__((aligned(8)));
+
+struct bpf_iter_kmem_cache_kern {
+	struct kmem_cache *pos;
+} __attribute__((aligned(8)));
+
+#define KMEM_CACHE_POS_START  ((void *)1L)
+
+__bpf_kfunc_start_defs();
+
+__bpf_kfunc int bpf_iter_kmem_cache_new(struct bpf_iter_kmem_cache *it)
+{
+	struct bpf_iter_kmem_cache_kern *kit = (void *)it;
+
+	BUILD_BUG_ON(sizeof(*kit) > sizeof(*it));
+	BUILD_BUG_ON(__alignof__(*kit) != __alignof__(*it));
+
+	kit->pos = KMEM_CACHE_POS_START;
+	return 0;
+}
+
+__bpf_kfunc struct kmem_cache *bpf_iter_kmem_cache_next(struct bpf_iter_kmem_cache *it)
+{
+	struct bpf_iter_kmem_cache_kern *kit = (void *)it;
+	struct kmem_cache *prev = kit->pos;
+	struct kmem_cache *next;
+	bool destroy = false;
+
+	if (!prev)
+		return NULL;
+
+	mutex_lock(&slab_mutex);
+
+	if (list_empty(&slab_caches)) {
+		mutex_unlock(&slab_mutex);
+		return NULL;
+	}
+
+	if (prev == KMEM_CACHE_POS_START)
+		next = list_first_entry(&slab_caches, struct kmem_cache, list);
+	else if (list_last_entry(&slab_caches, struct kmem_cache, list) == prev)
+		next = NULL;
+	else
+		next = list_next_entry(prev, list);
+
+	/* boot_caches have negative refcount, don't touch them */
+	if (next && next->refcount > 0)
+		next->refcount++;
+
+	/* Skip kmem_cache_destroy() for active entries */
+	if (prev && prev != KMEM_CACHE_POS_START) {
+		if (prev->refcount > 1)
+			prev->refcount--;
+		else if (prev->refcount == 1)
+			destroy = true;
+	}
+
+	mutex_unlock(&slab_mutex);
+
+	if (destroy)
+		kmem_cache_destroy(prev);
+
+	kit->pos = next;
+	return next;
+}
+
+__bpf_kfunc void bpf_iter_kmem_cache_destroy(struct bpf_iter_kmem_cache *it)
+{
+	struct bpf_iter_kmem_cache_kern *kit = (void *)it;
+	struct kmem_cache *s = kit->pos;
+	bool destroy = false;
+
+	if (s == NULL || s == KMEM_CACHE_POS_START)
+		return;
+
+	mutex_lock(&slab_mutex);
+
+	/* Skip kmem_cache_destroy() for active entries */
+	if (s->refcount > 1)
+		s->refcount--;
+	else if (s->refcount == 1)
+		destroy = true;
+
+	mutex_unlock(&slab_mutex);
+
+	if (destroy)
+		kmem_cache_destroy(s);
+}
+
+__bpf_kfunc_end_defs();
+
 struct bpf_iter__kmem_cache {
 	__bpf_md_ptr(struct bpf_iter_meta *, meta);
 	__bpf_md_ptr(struct kmem_cache *, s);
 };
 
+union kmem_cache_iter_priv {
+	struct bpf_iter_kmem_cache it;
+	struct bpf_iter_kmem_cache_kern kit;
+};
+
 static void *kmem_cache_iter_seq_start(struct seq_file *seq, loff_t *pos)
 {
 	loff_t cnt = 0;
 	bool found = false;
 	struct kmem_cache *s;
+	union kmem_cache_iter_priv *p = seq->private;
 
 	mutex_lock(&slab_mutex);
 
@@ -43,8 +143,9 @@ static void *kmem_cache_iter_seq_start(struct seq_file *seq, loff_t *pos)
 	mutex_unlock(&slab_mutex);
 
 	if (!found)
-		return NULL;
+		s = NULL;
 
+	p->kit.pos = s;
 	return s;
 }
 
@@ -55,63 +156,24 @@ static void kmem_cache_iter_seq_stop(struct seq_file *seq, void *v)
 		.meta = &meta,
 		.s = v,
 	};
+	union kmem_cache_iter_priv *p = seq->private;
 	struct bpf_prog *prog;
-	bool destroy = false;
 
 	meta.seq = seq;
 	prog = bpf_iter_get_info(&meta, true);
 	if (prog && !ctx.s)
 		bpf_iter_run_prog(prog, &ctx);
 
-	if (ctx.s == NULL)
-		return;
-
-	mutex_lock(&slab_mutex);
-
-	/* Skip kmem_cache_destroy() for active entries */
-	if (ctx.s->refcount > 1)
-		ctx.s->refcount--;
-	else if (ctx.s->refcount == 1)
-		destroy = true;
-
-	mutex_unlock(&slab_mutex);
-
-	if (destroy)
-		kmem_cache_destroy(ctx.s);
+	bpf_iter_kmem_cache_destroy(&p->it);
 }
 
 static void *kmem_cache_iter_seq_next(struct seq_file *seq, void *v, loff_t *pos)
 {
-	struct kmem_cache *s = v;
-	struct kmem_cache *next = NULL;
-	bool destroy = false;
+	union kmem_cache_iter_priv *p = seq->private;
 
 	++*pos;
 
-	mutex_lock(&slab_mutex);
-
-	if (list_last_entry(&slab_caches, struct kmem_cache, list) != s) {
-		next = list_next_entry(s, list);
-
-		WARN_ON_ONCE(next->refcount == 0);
-
-		/* boot_caches have negative refcount, don't touch them */
-		if (next->refcount > 0)
-			next->refcount++;
-	}
-
-	/* Skip kmem_cache_destroy() for active entries */
-	if (s->refcount > 1)
-		s->refcount--;
-	else if (s->refcount == 1)
-		destroy = true;
-
-	mutex_unlock(&slab_mutex);
-
-	if (destroy)
-		kmem_cache_destroy(s);
-
-	return next;
+	return bpf_iter_kmem_cache_next(&p->it);
 }
 
 static int kmem_cache_iter_seq_show(struct seq_file *seq, void *v)
@@ -143,6 +205,7 @@ BTF_ID_LIST_GLOBAL_SINGLE(bpf_kmem_cache_btf_id, struct, kmem_cache)
 
 static const struct bpf_iter_seq_info kmem_cache_iter_seq_info = {
 	.seq_ops		= &kmem_cache_iter_seq_ops,
+	.seq_priv_size		= sizeof(union kmem_cache_iter_priv),
 };
 
 static void bpf_iter_kmem_cache_show_fdinfo(const struct bpf_iter_aux_info *aux,
-- 
2.47.0.163.g1226f6d8fa-goog

Powered by blists - more mailing lists