BPF List
 help / color / mirror / Atom feed
From: Emil Tsalapatis <emil@etsalapatis.com>
To: bpf@vger.kernel.org
Cc: ast@kernel.org, andrii@kernel.org, eddyz87@gmail.com,
	memxor@gmail.com, daniel@iogearbox.net, netdev@vger.kernel.org,
	Emil Tsalapatis <emil@etsalapatis.com>,
	Nicholas Carlini <nicholas@carlini.com>
Subject: [PATCH bpf 07/11] bpf: Prevent variable arena/non-arena register contents
Date: Wed, 16 Sep 2026 05:08:25 +0000	[thread overview]
Message-ID: <20260916050830.8774-8-emil@etsalapatis.com> (raw)
In-Reply-To: <20260916050830.8774-1-emil@etsalapatis.com>

The verifier marks ALU instructions that include at least
one arena operand with needs_zext: These instructions are
fixed up after verification to be ALU32 instructions to
ensure that the result is a valid offset into an arena.
However, different code paths may provide two non-arena
64-bit arguments to the same instruction. The result of
the operation in that code path is wrong, since it is
now unexpectedly truncated to 32 bits and zero-extended.

Add logic to the verifier to ensure every instruction either
always has at least one PTR_TO_ARENA argument, or never does.
Since needs_zext already tracks the first scenario, add a
prevent_zext field in bpf_insn_aux to track the latter.
Reject instructions that use arena arguments and have prevent_zext
set, or do not have arena arguments and have needs_zext set.

Fixes: 6082b6c328b5 ("bpf: Recognize addr_space_cast instruction in the verifier.")
Reported-by: Nicholas Carlini <nicholas@carlini.com>
Suggested-by: Nicholas Carlini <nicholas@carlini.com>
Signed-off-by: Emil Tsalapatis <emil@etsalapatis.com>
---
 include/linux/bpf_verifier.h |  1 +
 kernel/bpf/verifier.c        | 24 +++++++++++++++++++++++-
 2 files changed, 24 insertions(+), 1 deletion(-)

diff --git a/include/linux/bpf_verifier.h b/include/linux/bpf_verifier.h
index cf85141ea..d5b4ab0ba 100644
--- a/include/linux/bpf_verifier.h
+++ b/include/linux/bpf_verifier.h
@@ -679,6 +679,7 @@ struct bpf_insn_aux_data {
 	bool nospec_result; /* result is unsafe under speculation, nospec must follow */
 	bool zext_dst; /* this insn zero extends dst reg */
 	bool needs_zext; /* alu op needs to clear upper bits */
+	bool prevent_zext; /* alu op cannot be zext (already used with 64-bit scalars) */
 	bool non_sleepable; /* helper/kfunc may be called from non-sleepable context */
 	bool is_iter_next; /* bpf_iter_<type>_next() kfunc call */
 	bool call_with_percpu_alloc_ptr; /* {this,per}_cpu_ptr() with prog percpu alloc */
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 507bc14b4..2c08ea94d 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -16056,6 +16056,7 @@ static int adjust_reg_min_max_vals(struct bpf_verifier_env *env,
 	struct bpf_reg_state *regs = state->regs, *dst_reg, *src_reg;
 	struct bpf_reg_state *ptr_reg = NULL, off_reg = {0};
 	bool alu32 = (BPF_CLASS(insn->code) != BPF_ALU64);
+	struct bpf_insn_aux_data *aux = cur_aux(env);
 	u8 opcode = BPF_OP(insn->code);
 	int err;
 
@@ -16067,12 +16068,23 @@ static int adjust_reg_min_max_vals(struct bpf_verifier_env *env,
 
 	/* Case where at least one operand is an arena. */
 	if (dst_reg->type == PTR_TO_ARENA || (src_reg && src_reg->type == PTR_TO_ARENA)) {
-		struct bpf_insn_aux_data *aux = cur_aux(env);
 
 		if (dst_reg->type != PTR_TO_ARENA)
 			*dst_reg = *src_reg;
 
 		if (BPF_CLASS(insn->code) == BPF_ALU64) {
+			/*
+			 * Only arena pointers set needs_zext, but doing so
+			 * modifies the instruction at fixup time to an ALU32
+			 * and makes it unsuitable for 64-bit scalar args. We
+			 * prevent zext from being set if the instruction has
+			 * been previously called with non-arena registers.
+			 */
+			if (aux->prevent_zext) {
+				verbose(env, "same insn cannot be used with and without arena pointer\n");
+				return -EINVAL;
+			}
+
 			/*
 			 * 32-bit operations zero upper bits automatically.
 			 * 64-bit operations need to be converted to 32.
@@ -16085,6 +16097,16 @@ static int adjust_reg_min_max_vals(struct bpf_verifier_env *env,
 		return 0;
 	}
 
+	/* Prevent the instruction from being used with arena pointers (see above). */
+	if (env->prog->aux->arena && BPF_CLASS(insn->code) == BPF_ALU64) {
+		if (aux->needs_zext) {
+			verbose(env, "same insn cannot be used with and without arena pointer\n");
+			return -EINVAL;
+		}
+
+		aux->prevent_zext = true;
+	}
+
 	if (dst_reg->type != SCALAR_VALUE)
 		ptr_reg = dst_reg;
 
-- 
2.54.0


  parent reply	other threads:[~2026-09-16  5:08 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-16  5:08 [PATCH bpf 00/11] skb/arena bugfixes Emil Tsalapatis
2026-09-16  5:08 ` [PATCH bpf 01/11] bpf: Fix bounds check for skb-backed dynptrs Emil Tsalapatis
2026-09-16 11:52   ` Jiayuan Chen
2026-09-16 23:08   ` Jakub Kicinski
2026-09-16  5:08 ` [PATCH bpf 02/11] selftests/bpf: Test dynptr slices past end of skb Emil Tsalapatis
2026-09-16  5:18   ` sashiko-bot
2026-09-16  5:08 ` [PATCH bpf 03/11] bpf: Fix bpf_sock context code generation Emil Tsalapatis
2026-09-16 12:21   ` Jiayuan Chen
2026-09-16  5:08 ` [PATCH bpf 04/11] selftests/bpf: Add selftests for rx_queue_mapping context access Emil Tsalapatis
2026-09-16  5:08 ` [PATCH bpf 05/11] bpf: Reject pkt arguments in mutating subprogs Emil Tsalapatis
2026-09-16  5:57   ` Amery Hung
2026-09-16  5:58     ` Amery Hung
2026-09-16 18:41     ` Emil Tsalapatis
2026-09-16 19:26       ` Amery Hung
2026-09-16  5:08 ` [PATCH bpf 06/11] selftests/bpf: Test rejection of pkt args to " Emil Tsalapatis
2026-09-16  5:08 ` Emil Tsalapatis [this message]
2026-09-16  5:08 ` [PATCH bpf 08/11] selftests/bpf: Test for mixed arena/nonarena code paths Emil Tsalapatis
2026-09-16  5:17   ` sashiko-bot
2026-09-16  5:08 ` [PATCH bpf 09/11] bpf: Track whether dynptr type is known Emil Tsalapatis
2026-09-16  5:22   ` sashiko-bot
2026-09-16 18:41     ` Emil Tsalapatis
2026-09-16  5:08 ` [PATCH bpf 10/11] bpf: Track skb memory invalidation by packet-backed dynptrs Emil Tsalapatis
2026-09-16  5:08 ` [PATCH bpf 11/11] selftests/bpf: Test dynptr slice invalidation on skb clobber Emil Tsalapatis

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=20260916050830.8774-8-emil@etsalapatis.com \
    --to=emil@etsalapatis.com \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=daniel@iogearbox.net \
    --cc=eddyz87@gmail.com \
    --cc=memxor@gmail.com \
    --cc=netdev@vger.kernel.org \
    --cc=nicholas@carlini.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