From: Puranjay Mohan <puranjay@kernel.org>
To: bpf@vger.kernel.org
Cc: Puranjay Mohan <puranjay@kernel.org>,
Puranjay Mohan <puranjay12@gmail.com>,
Alexei Starovoitov <ast@kernel.org>,
Andrii Nakryiko <andrii@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Martin KaFai Lau <martin.lau@kernel.org>,
Eduard Zingerman <eddyz87@gmail.com>,
Kumar Kartikeya Dwivedi <memxor@gmail.com>,
Mykyta Yatsenko <mykyta.yatsenko5@gmail.com>,
kernel-team@meta.com
Subject: [PATCH bpf-next 0/3] bpf: Improve linked register tracking
Date: Wed, 7 Jan 2026 12:39:33 -0800 [thread overview]
Message-ID: <20260107203941.1063754-1-puranjay@kernel.org> (raw)
This series extends the BPF verifier's linked register tracking to handle
negative offsets and BPF_SUB operations, enabling better bounds propagation for
common arithmetic patterns.
The verifier previously only tracked positive constant deltas between linked
registers using BPF_ADD. This meant patterns using negative offsets or
subtraction couldn't benefit from bounds propagation:
r1 = r0
r1 += -4
if r1 s>= 0 goto ... // r1 >= 0 implies r0 >= 4
// verifier couldn't propagate bounds back to r0
Patch 1 extends scalar_min_max_add() to:
- Accept BPF_SUB in addition to BPF_ADD (treating r1 -= 4 as r1 += -4)
- Change the overflow check to properly validate s32 range
- Add a guard against S32_MIN negation overflow
- Retain the !alu32 restriction due to known issues with 32-bit ALU upper bits
Patches 2-3 update the selftests:
- Patch 2 adds comprehensive tests covering success cases (negative offsets,
BPF_SUB), failure cases (32-bit ALU, double ADD), and large delta edge cases
(S32_MIN/S32_MAX offsets)
- Patch 3 updates an existing test's expected output to reflect the new
tracking behavior
Puranjay Mohan (3):
bpf: Support negative offsets and BPF_SUB for linked register tracking
selftests/bpf: Add tests for linked register tracking with negative
offsets
selftests/bpf: Update expected output for sub64_partial_overflow test
kernel/bpf/verifier.c | 26 ++-
.../selftests/bpf/progs/verifier_bounds.c | 2 +-
.../bpf/progs/verifier_linked_scalars.c | 213 ++++++++++++++++++
3 files changed, 233 insertions(+), 8 deletions(-)
base-commit: 2175ccfb93fd91d0ece74684eb7ab9443de806ec
--
2.47.3
next reply other threads:[~2026-01-07 20:40 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-07 20:39 Puranjay Mohan [this message]
2026-01-07 20:39 ` [PATCH bpf-next 1/3] bpf: Support negative offsets and BPF_SUB for linked register tracking Puranjay Mohan
2026-01-08 1:40 ` Eduard Zingerman
2026-01-08 1:47 ` Eduard Zingerman
2026-01-08 2:53 ` Alexei Starovoitov
2026-01-07 20:39 ` [PATCH bpf-next 2/3] selftests/bpf: Add tests for linked register tracking with negative offsets Puranjay Mohan
2026-01-08 2:11 ` Eduard Zingerman
2026-01-21 0:46 ` __description(). Was: " Alexei Starovoitov
2026-01-08 6:55 ` Eduard Zingerman
2026-01-08 11:33 ` Puranjay Mohan
2026-01-07 20:39 ` [PATCH bpf-next 3/3] selftests/bpf: Update expected output for sub64_partial_overflow test Puranjay Mohan
2026-01-08 6:59 ` Eduard Zingerman
2026-01-08 10:39 ` Puranjay Mohan
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=20260107203941.1063754-1-puranjay@kernel.org \
--to=puranjay@kernel.org \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=eddyz87@gmail.com \
--cc=kernel-team@meta.com \
--cc=martin.lau@kernel.org \
--cc=memxor@gmail.com \
--cc=mykyta.yatsenko5@gmail.com \
--cc=puranjay12@gmail.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