Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   =?UTF-8?q?Iago=20L=C3=B3pez=20Galeiras?= <iago@kinvolk.io>
To:     john.fastabend@gmail.com, ast@kernel.org, daniel@iogearbox.net
Cc:     alban@kinvolk.io, krzesimir@kinvolk.io, bpf@vger.kernel.org,
        netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH bpf-next v5 4/4] selftests: bpf: verifier: read netns_dev and netns_ino from struct bpf_sock_ops
Date:   Fri,  7 Jun 2019 16:11:06 +0200
Message-Id: <20190607141106.32148-5-iago@kinvolk.io>
In-Reply-To: <20190607141106.32148-1-iago@kinvolk.io>
References: <20190607141106.32148-1-iago@kinvolk.io>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

From: Alban Crequy <alban@kinvolk.io>

Tested with:
> $ sudo ./test_verifier
> ...
> #905/p sockops accessing bpf_sock_ops->netns_dev, ok OK
> #906/p sockops accessing bpf_sock_ops->netns_ino, ok OK
> ...
> Summary: 1421 PASSED, 0 SKIPPED, 0 FAILED

Signed-off-by: Alban Crequy <alban@kinvolk.io>

---

Changes since v1:
- This is a new selftest (review from Song)

Changes since v2:
- test partial reads on netns_dev (review from Y Song)
- split in two tests
---
 .../testing/selftests/bpf/verifier/var_off.c  | 53 +++++++++++++++++++
 1 file changed, 53 insertions(+)

diff --git a/tools/testing/selftests/bpf/verifier/var_off.c b/tools/testing/selftests/bpf/verifier/var_off.c
index 8504ac937809..9e4c6c78eb9d 100644
--- a/tools/testing/selftests/bpf/verifier/var_off.c
+++ b/tools/testing/selftests/bpf/verifier/var_off.c
@@ -246,3 +246,56 @@
 	.result = ACCEPT,
 	.prog_type = BPF_PROG_TYPE_LWT_IN,
 },
+{
+	"sockops accessing bpf_sock_ops->netns_dev, ok",
+	.insns = {
+	BPF_LDX_MEM(BPF_DW, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev)),
+
+	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev)),
+	BPF_LDX_MEM(BPF_W, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev) + 4),
+
+	BPF_LDX_MEM(BPF_H, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev)),
+	BPF_LDX_MEM(BPF_H, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev) + 2),
+	BPF_LDX_MEM(BPF_H, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev) + 4),
+	BPF_LDX_MEM(BPF_H, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev) + 6),
+
+	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev)),
+	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev) + 1),
+	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev) + 2),
+	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev) + 3),
+	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev) + 4),
+	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev) + 5),
+	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev) + 6),
+	BPF_LDX_MEM(BPF_B, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_dev) + 7),
+	BPF_MOV64_IMM(BPF_REG_0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.result = ACCEPT,
+	.prog_type = BPF_PROG_TYPE_SOCK_OPS,
+},
+{
+	"sockops accessing bpf_sock_ops->netns_ino, ok",
+	.insns = {
+	BPF_LDX_MEM(BPF_DW, BPF_REG_4, BPF_REG_1, offsetof(struct bpf_sock_ops,
+							   netns_ino)),
+	BPF_MOV64_IMM(BPF_REG_0, 0),
+	BPF_EXIT_INSN(),
+	},
+	.result = ACCEPT,
+	.prog_type = BPF_PROG_TYPE_SOCK_OPS,
+},
-- 
2.21.0