Merge branch 'writeable-bpf-tracepoints'

Matt Mullins says:

====================
This adds an opt-in interface for tracepoints to expose a writable context to
BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE programs that are attached, while
supporting read-only access from existing BPF_PROG_TYPE_RAW_TRACEPOINT
programs, as well as from non-BPF-based tracepoints.

The initial motivation is to support tracing that can be observed from the
remote end of an NBD socket, e.g. by adding flags to the struct nbd_request
header.  Earlier attempts included adding an NBD-specific tracepoint fd, but in
code review, I was recommended to implement it more generically -- as a result,
this patchset is far simpler than my initial try.

v4->v5:
  * rebased onto bpf-next/master and fixed merge conflicts
  * "tools: sync bpf.h" also syncs comments that have previously changed
    in bpf-next

v3->v4:
  * fixed a silly copy/paste typo in include/trace/events/bpf_test_run.h
    (_TRACE_NBD_H -> _TRACE_BPF_TEST_RUN_H)
  * fixed incorrect/misleading wording in patch 1's commit message,
    since the pointer cannot be directly dereferenced in a
    BPF_PROG_TYPE_RAW_TRACEPOINT
  * cleaned up the error message wording if the prog_tests fail
  * Addressed feedback from Yonghong
    * reject non-pointer-sized accesses to the buffer pointer
    * use sizeof(struct nbd_request) as one-byte-past-the-end in
      raw_tp_writable_reject_nbd_invalid.c
    * use BPF_MOV64_IMM instead of BPF_LD_IMM64

v2->v3:
  * Andrew addressed Josef's comments:
    * C-style commenting in nbd.c
    * Collapsed identical events into a single DECLARE_EVENT_CLASS.
      This saves about 2kB of kernel text

v1->v2:
  * add selftests
    * sync tools/include/uapi/linux/bpf.h
  * reject variable offset into the buffer
  * add string representation of PTR_TO_TP_BUFFER to reg_type_str
====================

Signed-off-by: Alexei Starovoitov <ast@kernel.org>

3 files changed, 156 insertions, 0 deletions

diff --git a/tools/testing/selftests/bpf/prog_tests/raw_tp_writable_reject_nbd_invalid.c b/tools/testing/selftests/bpf/prog_tests/raw_tp_writable_reject_nbd_invalid.c
new file mode 100644
index 000000000000..9807336a3016
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/raw_tp_writable_reject_nbd_invalid.c
@@ -0,0 +1,42 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <test_progs.h>
+#include <linux/nbd.h>
+
+void test_raw_tp_writable_reject_nbd_invalid(void)
+{
+	__u32 duration = 0;
+	char error[4096];
+	int bpf_fd = -1, tp_fd = -1;
+
+	const struct bpf_insn program[] = {
+		/* r6 is our tp buffer */
+		BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
+		/* one byte beyond the end of the nbd_request struct */
+		BPF_LDX_MEM(BPF_B, BPF_REG_0, BPF_REG_6,
+			    sizeof(struct nbd_request)),
+		BPF_EXIT_INSN(),
+	};
+
+	struct bpf_load_program_attr load_attr = {
+		.prog_type = BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE,
+		.license = "GPL v2",
+		.insns = program,
+		.insns_cnt = sizeof(program) / sizeof(struct bpf_insn),
+		.log_level = 2,
+	};
+
+	bpf_fd = bpf_load_program_xattr(&load_attr, error, sizeof(error));
+	if (CHECK(bpf_fd < 0, "bpf_raw_tracepoint_writable load",
+		  "failed: %d errno %d\n", bpf_fd, errno))
+		return;
+
+	tp_fd = bpf_raw_tracepoint_open("nbd_send_request", bpf_fd);
+	if (CHECK(tp_fd >= 0, "bpf_raw_tracepoint_writable open",
+		  "erroneously succeeded\n"))
+		goto out_bpffd;
+
+	close(tp_fd);
+out_bpffd:
+	close(bpf_fd);
+}
diff --git a/tools/testing/selftests/bpf/prog_tests/raw_tp_writable_test_run.c b/tools/testing/selftests/bpf/prog_tests/raw_tp_writable_test_run.c
new file mode 100644
index 000000000000..5c45424cac5f
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/raw_tp_writable_test_run.c
@@ -0,0 +1,80 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <test_progs.h>
+#include <linux/nbd.h>
+
+void test_raw_tp_writable_test_run(void)
+{
+	__u32 duration = 0;
+	char error[4096];
+
+	const struct bpf_insn trace_program[] = {
+		BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
+		BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_6, 0),
+		BPF_MOV64_IMM(BPF_REG_0, 42),
+		BPF_STX_MEM(BPF_W, BPF_REG_6, BPF_REG_0, 0),
+		BPF_EXIT_INSN(),
+	};
+
+	struct bpf_load_program_attr load_attr = {
+		.prog_type = BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE,
+		.license = "GPL v2",
+		.insns = trace_program,
+		.insns_cnt = sizeof(trace_program) / sizeof(struct bpf_insn),
+		.log_level = 2,
+	};
+
+	int bpf_fd = bpf_load_program_xattr(&load_attr, error, sizeof(error));
+	if (CHECK(bpf_fd < 0, "bpf_raw_tracepoint_writable loaded",
+		  "failed: %d errno %d\n", bpf_fd, errno))
+		return;
+
+	const struct bpf_insn skb_program[] = {
+		BPF_MOV64_IMM(BPF_REG_0, 0),
+		BPF_EXIT_INSN(),
+	};
+
+	struct bpf_load_program_attr skb_load_attr = {
+		.prog_type = BPF_PROG_TYPE_SOCKET_FILTER,
+		.license = "GPL v2",
+		.insns = skb_program,
+		.insns_cnt = sizeof(skb_program) / sizeof(struct bpf_insn),
+	};
+
+	int filter_fd =
+		bpf_load_program_xattr(&skb_load_attr, error, sizeof(error));
+	if (CHECK(filter_fd < 0, "test_program_loaded", "failed: %d errno %d\n",
+		  filter_fd, errno))
+		goto out_bpffd;
+
+	int tp_fd = bpf_raw_tracepoint_open("bpf_test_finish", bpf_fd);
+	if (CHECK(tp_fd < 0, "bpf_raw_tracepoint_writable opened",
+		  "failed: %d errno %d\n", tp_fd, errno))
+		goto out_filterfd;
+
+	char test_skb[128] = {
+		0,
+	};
+
+	__u32 prog_ret;
+	int err = bpf_prog_test_run(filter_fd, 1, test_skb, sizeof(test_skb), 0,
+				    0, &prog_ret, 0);
+	CHECK(err != 42, "test_run",
+	      "tracepoint did not modify return value\n");
+	CHECK(prog_ret != 0, "test_run_ret",
+	      "socket_filter did not return 0\n");
+
+	close(tp_fd);
+
+	err = bpf_prog_test_run(filter_fd, 1, test_skb, sizeof(test_skb), 0, 0,
+				&prog_ret, 0);
+	CHECK(err != 0, "test_run_notrace",
+	      "test_run failed with %d errno %d\n", err, errno);
+	CHECK(prog_ret != 0, "test_run_ret_notrace",
+	      "socket_filter did not return 0\n");
+
+out_filterfd:
+	close(filter_fd);
+out_bpffd:
+	close(bpf_fd);
+}
diff --git a/tools/testing/selftests/bpf/verifier/raw_tp_writable.c b/tools/testing/selftests/bpf/verifier/raw_tp_writable.c
new file mode 100644
index 000000000000..95b5d70a1dc1
--- /dev/null
+++ b/tools/testing/selftests/bpf/verifier/raw_tp_writable.c
@@ -0,0 +1,34 @@
+{
+	"raw_tracepoint_writable: reject variable offset",
+	.insns = {
+		/* r6 is our tp buffer */
+		BPF_LDX_MEM(BPF_DW, BPF_REG_6, BPF_REG_1, 0),
+
+		BPF_LD_MAP_FD(BPF_REG_1, 0),
+		/* move the key (== 0) to r10-8 */
+		BPF_MOV32_IMM(BPF_REG_0, 0),
+		BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
+		BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -8),
+		BPF_STX_MEM(BPF_DW, BPF_REG_2, BPF_REG_0, 0),
+		/* lookup in the map */
+		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0,
+			     BPF_FUNC_map_lookup_elem),
+
+		/* exit clean if null */
+		BPF_JMP_IMM(BPF_JNE, BPF_REG_0, 0, 1),
+		BPF_EXIT_INSN(),
+
+		/* shift the buffer pointer to a variable location */
+		BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_0, 0),
+		BPF_ALU64_REG(BPF_ADD, BPF_REG_6, BPF_REG_0),
+		/* clobber whatever's there */
+		BPF_MOV64_IMM(BPF_REG_7, 4242),
+		BPF_STX_MEM(BPF_DW, BPF_REG_6, BPF_REG_7, 0),
+
+		BPF_MOV64_IMM(BPF_REG_0, 0),
+		BPF_EXIT_INSN(),
+	},
+	.fixup_map_hash_8b = { 1, },
+	.prog_type = BPF_PROG_TYPE_RAW_TRACEPOINT_WRITABLE,
+	.errstr = "R6 invalid variable buffer offset: off=0, var_off=(0x0; 0xffffffff)",
+},