BPF List
 help / color / mirror / Atom feed
From: Kumar Kartikeya Dwivedi <memxor@gmail.com>
To: bpf@vger.kernel.org
Cc: "Alexei Starovoitov" <ast@kernel.org>,
	"Andrii Nakryiko" <andrii@kernel.org>,
	"Daniel Borkmann" <daniel@iogearbox.net>,
	"Toke Høiland-Jørgensen" <toke@redhat.com>,
	"Jesper Dangaard Brouer" <brouer@redhat.com>
Subject: [PATCH bpf-next v2 06/15] bpf: Allow storing user kptr in map
Date: Thu, 17 Mar 2022 17:29:48 +0530	[thread overview]
Message-ID: <20220317115957.3193097-7-memxor@gmail.com> (raw)
In-Reply-To: <20220317115957.3193097-1-memxor@gmail.com>

Recently, verifier gained __user annotation support [0] where it
prevents BPF program from normally derefering user memory pointer in the
kernel, and instead requires use of bpf_probe_read_user. We can allow
the user to also store these pointers in BPF maps, with the logic that
whenever user loads it from the BPF map, it gets marked as MEM_USER. The
tag 'kptr_user' is used to tag such pointers.

  [0]: https://lore.kernel.org/bpf/20220127154555.650886-1-yhs@fb.com

Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
---
 include/linux/bpf.h   |  1 +
 kernel/bpf/btf.c      | 13 ++++++++++---
 kernel/bpf/verifier.c | 15 ++++++++++++---
 3 files changed, 23 insertions(+), 6 deletions(-)

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 433f5cb161cf..989f47334215 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -163,6 +163,7 @@ enum {
 enum {
 	BPF_MAP_VALUE_OFF_F_REF    = (1U << 0),
 	BPF_MAP_VALUE_OFF_F_PERCPU = (1U << 1),
+	BPF_MAP_VALUE_OFF_F_USER   = (1U << 2),
 };
 
 struct bpf_map_value_off_desc {
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 04d604931f59..12a89e55e77b 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3197,7 +3197,7 @@ static int btf_find_field_kptr(const struct btf *btf, const struct btf_type *t,
 			       u32 off, int sz, struct btf_field_info *info,
 			       int info_cnt, int idx)
 {
-	bool kptr_tag = false, kptr_ref_tag = false, kptr_percpu_tag = false;
+	bool kptr_tag = false, kptr_ref_tag = false, kptr_percpu_tag = false, kptr_user_tag = false;
 	int tags;
 
 	/* For PTR, sz is always == 8 */
@@ -3221,12 +3221,17 @@ static int btf_find_field_kptr(const struct btf *btf, const struct btf_type *t,
 			if (kptr_percpu_tag)
 				return -EEXIST;
 			kptr_percpu_tag = true;
+		} else if (!strcmp("kptr_user", __btf_name_by_offset(btf, t->name_off))) {
+			/* repeated tag */
+			if (kptr_user_tag)
+				return -EEXIST;
+			kptr_user_tag = true;
 		}
 		/* Look for next tag */
 		t = btf_type_by_id(btf, t->type);
 	}
 
-	tags = kptr_tag + kptr_ref_tag + kptr_percpu_tag;
+	tags = kptr_tag + kptr_ref_tag + kptr_percpu_tag + kptr_user_tag;
 	if (!tags)
 		return BTF_FIELD_IGNORE;
 	else if (tags > 1)
@@ -3241,7 +3246,9 @@ static int btf_find_field_kptr(const struct btf *btf, const struct btf_type *t,
 
 	if (idx >= info_cnt)
 		return -E2BIG;
-	if (kptr_percpu_tag)
+	if (kptr_user_tag)
+		info[idx].flags = BPF_MAP_VALUE_OFF_F_USER;
+	else if (kptr_percpu_tag)
 		info[idx].flags = BPF_MAP_VALUE_OFF_F_PERCPU;
 	else if (kptr_ref_tag)
 		info[idx].flags = BPF_MAP_VALUE_OFF_F_REF;
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index cc8f7250e43e..5325cc37797a 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -3521,7 +3521,11 @@ static int map_kptr_match_type(struct bpf_verifier_env *env,
 	const char *reg_name = "";
 	bool fixed_off_ok = true;
 
-	if (off_desc->flags & BPF_MAP_VALUE_OFF_F_PERCPU) {
+	if (off_desc->flags & BPF_MAP_VALUE_OFF_F_USER) {
+		if (reg->type != (PTR_TO_BTF_ID | MEM_USER) &&
+		    reg->type != (PTR_TO_BTF_ID | PTR_MAYBE_NULL | MEM_USER))
+			goto bad_type;
+	} else if (off_desc->flags & BPF_MAP_VALUE_OFF_F_PERCPU) {
 		if (reg->type != (PTR_TO_BTF_ID | MEM_PERCPU) &&
 		    reg->type != (PTR_TO_BTF_ID | PTR_MAYBE_NULL | MEM_PERCPU))
 			goto bad_type;
@@ -3565,7 +3569,9 @@ static int map_kptr_match_type(struct bpf_verifier_env *env,
 		goto bad_type;
 	return 0;
 bad_type:
-	if (off_desc->flags & BPF_MAP_VALUE_OFF_F_PERCPU)
+	if (off_desc->flags & BPF_MAP_VALUE_OFF_F_USER)
+		reg_type = PTR_TO_BTF_ID | PTR_MAYBE_NULL | MEM_USER;
+	else if (off_desc->flags & BPF_MAP_VALUE_OFF_F_PERCPU)
 		reg_type = PTR_TO_BTF_ID | PTR_MAYBE_NULL | MEM_PERCPU;
 	else
 		reg_type = PTR_TO_BTF_ID | PTR_MAYBE_NULL;
@@ -3583,9 +3589,9 @@ static int check_map_kptr_access(struct bpf_verifier_env *env, u32 regno,
 				 enum bpf_access_type t, int insn_idx)
 {
 	struct bpf_reg_state *reg = reg_state(env, regno), *val_reg;
+	bool ref_ptr = false, percpu_ptr = false, user_ptr = false;
 	struct bpf_insn *insn = &env->prog->insnsi[insn_idx];
 	enum bpf_type_flag reg_flags = PTR_MAYBE_NULL;
-	bool ref_ptr = false, percpu_ptr = false;
 	struct bpf_map_value_off_desc *off_desc;
 	int insn_class = BPF_CLASS(insn->code);
 	struct bpf_map *map = reg->map_ptr;
@@ -3615,8 +3621,11 @@ static int check_map_kptr_access(struct bpf_verifier_env *env, u32 regno,
 
 	ref_ptr = off_desc->flags & BPF_MAP_VALUE_OFF_F_REF;
 	percpu_ptr = off_desc->flags & BPF_MAP_VALUE_OFF_F_PERCPU;
+	user_ptr = off_desc->flags & BPF_MAP_VALUE_OFF_F_USER;
 	if (percpu_ptr)
 		reg_flags |= MEM_PERCPU;
+	else if (user_ptr)
+		reg_flags |= MEM_USER;
 
 	if (insn_class == BPF_LDX) {
 		if (WARN_ON_ONCE(value_regno < 0))
-- 
2.35.1


  parent reply	other threads:[~2022-03-17 12:00 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-17 11:59 [PATCH bpf-next v2 00/15] Introduce typed pointer support in BPF maps Kumar Kartikeya Dwivedi
2022-03-17 11:59 ` [PATCH bpf-next v2 01/15] bpf: Factor out fd returning from bpf_btf_find_by_name_kind Kumar Kartikeya Dwivedi
2022-03-17 11:59 ` [PATCH bpf-next v2 02/15] bpf: Make btf_find_field more generic Kumar Kartikeya Dwivedi
2022-03-19 17:55   ` Alexei Starovoitov
2022-03-19 19:31     ` Kumar Kartikeya Dwivedi
2022-03-19 20:06       ` Kumar Kartikeya Dwivedi
2022-03-19 21:30         ` Alexei Starovoitov
2022-03-17 11:59 ` [PATCH bpf-next v2 03/15] bpf: Allow storing unreferenced kptr in map Kumar Kartikeya Dwivedi
2022-03-19 18:15   ` Alexei Starovoitov
2022-03-19 18:52     ` Kumar Kartikeya Dwivedi
2022-03-19 21:17       ` Alexei Starovoitov
2022-03-19 21:39         ` Kumar Kartikeya Dwivedi
2022-03-19 21:50           ` Kumar Kartikeya Dwivedi
2022-03-19 22:57             ` Alexei Starovoitov
2022-03-17 11:59 ` [PATCH bpf-next v2 04/15] bpf: Allow storing referenced " Kumar Kartikeya Dwivedi
2022-03-19 18:24   ` Alexei Starovoitov
2022-03-19 18:59     ` Kumar Kartikeya Dwivedi
2022-03-19 21:23       ` Alexei Starovoitov
2022-03-19 21:43         ` Kumar Kartikeya Dwivedi
2022-03-20  0:57           ` Kumar Kartikeya Dwivedi
2022-03-17 11:59 ` [PATCH bpf-next v2 05/15] bpf: Allow storing percpu " Kumar Kartikeya Dwivedi
2022-03-19 18:30   ` Alexei Starovoitov
2022-03-19 19:04     ` Kumar Kartikeya Dwivedi
2022-03-19 21:26       ` Alexei Starovoitov
2022-03-19 21:45         ` Kumar Kartikeya Dwivedi
2022-03-19 23:01           ` Alexei Starovoitov
2022-03-17 11:59 ` Kumar Kartikeya Dwivedi [this message]
2022-03-19 18:28   ` [PATCH bpf-next v2 06/15] bpf: Allow storing user " Alexei Starovoitov
2022-03-19 19:02     ` Kumar Kartikeya Dwivedi
2022-03-17 11:59 ` [PATCH bpf-next v2 07/15] bpf: Prevent escaping of kptr loaded from maps Kumar Kartikeya Dwivedi
2022-03-17 11:59 ` [PATCH bpf-next v2 08/15] bpf: Adapt copy_map_value for multiple offset case Kumar Kartikeya Dwivedi
2022-03-19 18:34   ` Alexei Starovoitov
2022-03-19 19:06     ` Kumar Kartikeya Dwivedi
2022-03-17 11:59 ` [PATCH bpf-next v2 09/15] bpf: Always raise reference in btf_get_module_btf Kumar Kartikeya Dwivedi
2022-03-19 18:43   ` Alexei Starovoitov
2022-03-17 11:59 ` [PATCH bpf-next v2 10/15] bpf: Populate pairs of btf_id and destructor kfunc in btf Kumar Kartikeya Dwivedi
2022-03-17 11:59 ` [PATCH bpf-next v2 11/15] bpf: Wire up freeing of referenced kptr Kumar Kartikeya Dwivedi
2022-03-17 11:59 ` [PATCH bpf-next v2 12/15] bpf: Teach verifier about kptr_get kfunc helpers Kumar Kartikeya Dwivedi
2022-03-17 11:59 ` [PATCH bpf-next v2 13/15] libbpf: Add kptr type tag macros to bpf_helpers.h Kumar Kartikeya Dwivedi
2022-03-17 11:59 ` [PATCH bpf-next v2 14/15] selftests/bpf: Add C tests for kptr Kumar Kartikeya Dwivedi
2022-03-17 11:59 ` [PATCH bpf-next v2 15/15] selftests/bpf: Add verifier " Kumar Kartikeya Dwivedi
2022-03-19 18:50 ` [PATCH bpf-next v2 00/15] Introduce typed pointer support in BPF maps patchwork-bot+netdevbpf

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220317115957.3193097-7-memxor@gmail.com \
    --to=memxor@gmail.com \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=brouer@redhat.com \
    --cc=daniel@iogearbox.net \
    --cc=toke@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox