[bpf-next,v10,04/24] bpf: Populate field_offs for inner_map_meta

Commit Message

Kumar Kartikeya Dwivedi Nov. 18, 2022, 1:55 a.m. UTC

Far too much code simply assumes that both btf_record and btf_field_offs
are set to valid pointers together, or both are unset. They go together
hand in hand as btf_record describes the special fields and
btf_field_offs is compact representation for runtime copying/zeroing.

It is very difficult to make this clear in the code when the only
exception to this universal invariant is inner_map_meta which is used
as reg->map_ptr in the verifier. This is simply a bug waiting to happen,
as in verifier context we cannot easily distinguish if PTR_TO_MAP_VALUE
is coming from an inner map, and if we ever end up using field_offs for
any reason in the future, we will silently ignore the special fields for
inner map case (as NULL is not an error but unset field_offs).

Hence, simply copy field_offs from inner map together with btf_record.

While at it, refactor code to unwind properly on errors with gotos.

Signed-off-by: Kumar Kartikeya Dwivedi <[e-mail]>
---
 kernel/bpf/map_in_map.c | 44 ++++++++++++++++++++++++++++++-----------
 1 file changed, 32 insertions(+), 12 deletions(-)

Patch

diff --git a/kernel/bpf/map_in_map.c b/kernel/bpf/map_in_map.c
index a423130a8720..fae6a6c33e2d 100644
--- a/kernel/bpf/map_in_map.c
+++ b/kernel/bpf/map_in_map.c
@@ -12,6 +12,7 @@  struct bpf_map *bpf_map_meta_alloc(int inner_map_ufd)
 	struct bpf_map *inner_map, *inner_map_meta;
 	u32 inner_map_meta_size;
 	struct fd f;
+	int ret;
 
 	f = fdget(inner_map_ufd);
 	inner_map = __bpf_map_get(f);
@@ -20,18 +21,18 @@  struct bpf_map *bpf_map_meta_alloc(int inner_map_ufd)
 
 	/* Does not support >1 level map-in-map */
 	if (inner_map->inner_map_meta) {
-		fdput(f);
-		return ERR_PTR(-EINVAL);
+		ret = -EINVAL;
+		goto put;
 	}
 
 	if (!inner_map->ops->map_meta_equal) {
-		fdput(f);
-		return ERR_PTR(-ENOTSUPP);
+		ret = -ENOTSUPP;
+		goto put;
 	}
 
 	if (btf_record_has_field(inner_map->record, BPF_SPIN_LOCK)) {
-		fdput(f);
-		return ERR_PTR(-ENOTSUPP);
+		ret = -ENOTSUPP;
+		goto put;
 	}
 
 	inner_map_meta_size = sizeof(*inner_map_meta);
@@ -41,8 +42,8 @@  struct bpf_map *bpf_map_meta_alloc(int inner_map_ufd)
 
 	inner_map_meta = kzalloc(inner_map_meta_size, GFP_USER);
 	if (!inner_map_meta) {
-		fdput(f);
-		return ERR_PTR(-ENOMEM);
+		ret = -ENOMEM;
+		goto put;
 	}
 
 	inner_map_meta->map_type = inner_map->map_type;
@@ -50,16 +51,27 @@  struct bpf_map *bpf_map_meta_alloc(int inner_map_ufd)
 	inner_map_meta->value_size = inner_map->value_size;
 	inner_map_meta->map_flags = inner_map->map_flags;
 	inner_map_meta->max_entries = inner_map->max_entries;
+
 	inner_map_meta->record = btf_record_dup(inner_map->record);
 	if (IS_ERR(inner_map_meta->record)) {
-		struct bpf_map *err_ptr = ERR_CAST(inner_map_meta->record);
 		/* btf_record_dup returns NULL or valid pointer in case of
 		 * invalid/empty/valid, but ERR_PTR in case of errors. During
 		 * equality NULL or IS_ERR is equivalent.
 		 */
-		kfree(inner_map_meta);
-		fdput(f);
-		return err_ptr;
+		ret = PTR_ERR(inner_map_meta->record);
+		goto free;
+	}
+	if (inner_map_meta->record) {
+		struct btf_field_offs *field_offs;
+		/* If btf_record is !IS_ERR_OR_NULL, then field_offs is always
+		 * valid.
+		 */
+		field_offs = kmemdup(inner_map->field_offs, sizeof(*inner_map->field_offs), GFP_KERNEL | __GFP_NOWARN);
+		if (!field_offs) {
+			ret = -ENOMEM;
+			goto free_rec;
+		}
+		inner_map_meta->field_offs = field_offs;
 	}
 	if (inner_map->btf) {
 		btf_get(inner_map->btf);
@@ -76,10 +88,18 @@  struct bpf_map *bpf_map_meta_alloc(int inner_map_ufd)
 
 	fdput(f);
 	return inner_map_meta;
+free_rec:
+	btf_record_free(inner_map_meta->record);
+free:
+	kfree(inner_map_meta);
+put:
+	fdput(f);
+	return ERR_PTR(ret);
 }
 
 void bpf_map_meta_free(struct bpf_map *map_meta)
 {
+	kfree(map_meta->field_offs);
 	bpf_map_free_record(map_meta);
 	btf_put(map_meta->btf);
 	kfree(map_meta);