From: Yonghong Song <yonghong.song@linux.dev>
To: bpf@vger.kernel.org
Cc: Alexei Starovoitov <ast@kernel.org>,
Andrii Nakryiko <andrii@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
"Jose E . Marchesi" <jose.marchesi@oracle.com>,
kernel-team@fb.com, Martin KaFai Lau <martin.lau@kernel.org>
Subject: [PATCH bpf-next v6 09/17] bpf: Track r11 registers in const_fold and liveness
Date: Sun, 19 Apr 2026 09:34:02 -0700 [thread overview]
Message-ID: <20260419163402.734987-1-yonghong.song@linux.dev> (raw)
In-Reply-To: <20260419163316.731019-1-yonghong.song@linux.dev>
Since BPF_REG_PARAMS (r11) is beyond MAX_BPF_REG (r11), using it as an
array index into the register tracking arrays in const_fold.c and
liveness.c could cause out-of-bounds accesses.
When dst_reg is BPF_REG_PARAMS (BPF_ST/BPF_STX storing to the stack
arg area), no tracked register state changes, so an early return is
sufficient.
When src_reg is BPF_REG_PARAMS (BPF_LDX loading from the stack arg
area), dst_reg is a normal register that gets overwritten. Simply
returning early would leave dst_reg with stale state from a prior
path, which could cause incorrect constant folding or liveness
analysis. Mark dst_reg as unknown/none before returning to prevent
this.
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
---
kernel/bpf/const_fold.c | 14 ++++++++++++--
kernel/bpf/liveness.c | 14 ++++++++++++--
2 files changed, 24 insertions(+), 4 deletions(-)
diff --git a/kernel/bpf/const_fold.c b/kernel/bpf/const_fold.c
index db73c4740b1e..41e5a406432c 100644
--- a/kernel/bpf/const_fold.c
+++ b/kernel/bpf/const_fold.c
@@ -51,13 +51,23 @@ static void const_reg_xfer(struct bpf_verifier_env *env, struct const_arg_info *
struct bpf_insn *insn, struct bpf_insn *insns, int idx)
{
struct const_arg_info unknown = { .state = CONST_ARG_UNKNOWN, .val = 0 };
- struct const_arg_info *dst = &ci_out[insn->dst_reg];
- struct const_arg_info *src = &ci_out[insn->src_reg];
+ struct const_arg_info *dst, *src;
u8 class = BPF_CLASS(insn->code);
u8 mode = BPF_MODE(insn->code);
u8 opcode = BPF_OP(insn->code) | BPF_SRC(insn->code);
int r;
+ /* Stack arguments using BPF_REG_PARAMS are outside the tracked register set. */
+ if (insn->dst_reg >= MAX_BPF_REG)
+ return;
+ if (insn->src_reg >= MAX_BPF_REG) {
+ if (class == BPF_LDX)
+ ci_out[insn->dst_reg] = unknown;
+ return;
+ }
+
+ dst = &ci_out[insn->dst_reg];
+ src = &ci_out[insn->src_reg];
switch (class) {
case BPF_ALU:
case BPF_ALU64:
diff --git a/kernel/bpf/liveness.c b/kernel/bpf/liveness.c
index 332e6e003f27..87022da94f3a 100644
--- a/kernel/bpf/liveness.c
+++ b/kernel/bpf/liveness.c
@@ -1068,11 +1068,21 @@ static void arg_track_xfer(struct bpf_verifier_env *env, struct bpf_insn *insn,
int depth = instance->depth;
u8 class = BPF_CLASS(insn->code);
u8 code = BPF_OP(insn->code);
- struct arg_track *dst = &at_out[insn->dst_reg];
- struct arg_track *src = &at_out[insn->src_reg];
+ struct arg_track *dst, *src;
struct arg_track none = { .frame = ARG_NONE };
int r;
+ /* Stack arguments using BPF_REG_PARAMS are outside the tracked register set. */
+ if (insn->dst_reg >= MAX_BPF_REG)
+ return;
+ if (insn->src_reg >= MAX_BPF_REG) {
+ if (class == BPF_LDX)
+ at_out[insn->dst_reg] = none;
+ return;
+ }
+
+ dst = &at_out[insn->dst_reg];
+ src = &at_out[insn->src_reg];
if (class == BPF_ALU64 && BPF_SRC(insn->code) == BPF_K) {
if (code == BPF_MOV) {
*dst = none;
--
2.52.0
next prev parent reply other threads:[~2026-04-19 16:34 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-04-19 16:33 [PATCH bpf-next v6 00/17] bpf: Support stack arguments for BPF functions and kfuncs Yonghong Song
2026-04-19 16:33 ` [PATCH bpf-next v6 01/17] bpf: Remove unused parameter from check_map_kptr_access() Yonghong Song
2026-04-19 16:33 ` [PATCH bpf-next v6 02/17] bpf: Refactor to avoid redundant calculation of bpf_reg_state Yonghong Song
2026-04-19 16:33 ` [PATCH bpf-next v6 03/17] bpf: Refactor to handle memory and size together Yonghong Song
2026-04-20 23:58 ` Alexei Starovoitov
2026-04-21 4:04 ` Yonghong Song
2026-04-19 16:33 ` [PATCH bpf-next v6 04/17] bpf: Prepare verifier logs for upcoming kfunc stack arguments Yonghong Song
2026-04-21 0:03 ` Alexei Starovoitov
2026-04-21 4:06 ` Yonghong Song
2026-04-21 6:07 ` Yonghong Song
2026-04-21 13:48 ` Alexei Starovoitov
2026-04-21 15:41 ` Yonghong Song
2026-04-21 15:46 ` Alexei Starovoitov
2026-04-21 16:37 ` Yonghong Song
2026-04-21 17:24 ` Yonghong Song
2026-04-19 16:33 ` [PATCH bpf-next v6 05/17] bpf: Introduce bpf register BPF_REG_PARAMS Yonghong Song
2026-04-19 17:06 ` sashiko-bot
2026-04-19 18:14 ` Yonghong Song
2026-04-19 16:33 ` [PATCH bpf-next v6 06/17] bpf: Reuse MAX_BPF_FUNC_ARGS for maximum number of arguments Yonghong Song
2026-04-19 16:33 ` [PATCH bpf-next v6 07/17] bpf: Support stack arguments for bpf functions Yonghong Song
2026-04-19 19:15 ` sashiko-bot
2026-04-20 4:35 ` Yonghong Song
2026-04-21 0:37 ` Alexei Starovoitov
2026-04-21 4:15 ` Yonghong Song
2026-04-19 16:33 ` [PATCH bpf-next v6 08/17] bpf: Reject stack arguments in non-JITed programs Yonghong Song
2026-04-19 18:21 ` sashiko-bot
2026-04-20 4:23 ` Yonghong Song
2026-04-19 16:34 ` Yonghong Song [this message]
2026-04-19 16:34 ` [PATCH bpf-next v6 10/17] bpf: Prepare architecture JIT support for stack arguments Yonghong Song
2026-04-19 16:34 ` [PATCH bpf-next v6 11/17] bpf: Enable r11 based insns Yonghong Song
2026-04-19 16:34 ` [PATCH bpf-next v6 12/17] bpf: Support stack arguments for kfunc calls Yonghong Song
2026-04-19 17:08 ` sashiko-bot
2026-04-19 18:18 ` Yonghong Song
2026-04-19 16:34 ` [PATCH bpf-next v6 13/17] bpf: Reject stack arguments if tail call reachable Yonghong Song
2026-04-19 17:08 ` sashiko-bot
2026-04-19 18:20 ` Yonghong Song
2026-04-19 16:34 ` [PATCH bpf-next v6 14/17] bpf,x86: Implement JIT support for stack arguments Yonghong Song
2026-04-19 17:25 ` sashiko-bot
2026-04-19 18:55 ` Yonghong Song
2026-04-19 16:34 ` [PATCH bpf-next v6 15/17] selftests/bpf: Add tests for BPF function " Yonghong Song
2026-04-19 17:15 ` sashiko-bot
2026-04-20 5:52 ` Yonghong Song
2026-04-19 16:34 ` [PATCH bpf-next v6 16/17] selftests/bpf: Add tests for stack argument validation Yonghong Song
2026-04-19 16:34 ` [PATCH bpf-next v6 17/17] selftests/bpf: Add verifier " Yonghong Song
2026-04-19 17:21 ` sashiko-bot
2026-04-20 6:14 ` Yonghong Song
2026-04-20 15:41 ` [PATCH bpf-next v6 00/17] bpf: Support stack arguments for BPF functions and kfuncs Puranjay Mohan
2026-04-20 20:22 ` Yonghong Song
2026-04-20 20:25 ` Puranjay Mohan
2026-04-20 21:49 ` Alexei Starovoitov
2026-04-20 23:44 ` Yonghong Song
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=20260419163402.734987-1-yonghong.song@linux.dev \
--to=yonghong.song@linux.dev \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=jose.marchesi@oracle.com \
--cc=kernel-team@fb.com \
--cc=martin.lau@kernel.org \
/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