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>,
kernel-team@fb.com, Martin KaFai Lau <martin.lau@kernel.org>
Subject: [PATCH bpf 3/3] selftests/bpf: Add a few tests to cover
Date: Sat, 15 Jun 2024 10:46:37 -0700 [thread overview]
Message-ID: <20240615174637.3995589-1-yonghong.song@linux.dev> (raw)
In-Reply-To: <20240615174621.3994321-1-yonghong.song@linux.dev>
Add three unit tests in verifier_movsx.c to cover
cases where missed var_off setting can cause
unexpected verification success or failure.
Signed-off-by: Yonghong Song <yonghong.song@linux.dev>
---
.../selftests/bpf/progs/verifier_movsx.c | 63 +++++++++++++++++++
1 file changed, 63 insertions(+)
diff --git a/tools/testing/selftests/bpf/progs/verifier_movsx.c b/tools/testing/selftests/bpf/progs/verifier_movsx.c
index cbb9d6714f53..028ec855587b 100644
--- a/tools/testing/selftests/bpf/progs/verifier_movsx.c
+++ b/tools/testing/selftests/bpf/progs/verifier_movsx.c
@@ -224,6 +224,69 @@ l0_%=: \
: __clobber_all);
}
+SEC("socket")
+__description("MOV32SX, S8, var_off u32_max")
+__failure __msg("infinite loop detected")
+__failure_unpriv __msg_unpriv("back-edge from insn 2 to 0")
+__naked void mov64sx_s32_varoff_1(void)
+{
+ asm volatile (" \
+l0_%=: \
+ r3 = *(u8 *)(r10 -387); \
+ w7 = (s8)w3; \
+ if w7 >= 0x2533823b goto l0_%=; \
+ w0 = 0; \
+ exit; \
+" :
+ :
+ : __clobber_all);
+}
+
+SEC("socket")
+__description("MOV32SX, S8, var_off not u32_max, positive after s8 extension")
+__success __retval(0)
+__failure_unpriv __msg_unpriv("frame pointer is read only")
+__naked void mov64sx_s32_varoff_2(void)
+{
+ asm volatile (" \
+ call %[bpf_get_prandom_u32]; \
+ r3 = r0; \
+ r3 &= 0xf; \
+ w7 = (s8)w3; \
+ if w7 s>= 16 goto l0_%=; \
+ w0 = 0; \
+ exit; \
+l0_%=: \
+ r10 = 1; \
+ exit; \
+" :
+ : __imm(bpf_get_prandom_u32)
+ : __clobber_all);
+}
+
+SEC("socket")
+__description("MOV32SX, S8, var_off not u32_max, negative after s8 extension")
+__success __retval(0)
+__failure_unpriv __msg_unpriv("frame pointer is read only")
+__naked void mov64sx_s32_varoff_3(void)
+{
+ asm volatile (" \
+ call %[bpf_get_prandom_u32]; \
+ r3 = r0; \
+ r3 &= 0xf; \
+ r3 |= 0x80; \
+ w7 = (s8)w3; \
+ if w7 s>= -5 goto l0_%=; \
+ w0 = 0; \
+ exit; \
+l0_%=: \
+ r10 = 1; \
+ exit; \
+" :
+ : __imm(bpf_get_prandom_u32)
+ : __clobber_all);
+}
+
#else
SEC("socket")
--
2.43.0
next prev parent reply other threads:[~2024-06-15 17:46 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-06-15 17:46 [PATCH bpf 0/3] bpf: Fix missed var_off related to movsx in verifier Yonghong Song
2024-06-15 17:46 ` [PATCH bpf 1/3] bpf: Add missed var_off setting in set_sext32_default_val() Yonghong Song
2024-06-15 17:46 ` [PATCH bpf 2/3] bpf: Add missed var_off setting in coerce_subreg_to_size_sx() Yonghong Song
2024-06-15 17:46 ` Yonghong Song [this message]
2024-06-17 17:42 ` [PATCH bpf 0/3] bpf: Fix missed var_off related to movsx in verifier Eduard Zingerman
2024-06-17 17:50 ` 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=20240615174637.3995589-1-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=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