bpf: remove useless version check for prog load

Existing libraries and tracing frameworks work around this kernel
version check by automatically deriving the kernel version from
uname(3) or similar such that the user does not need to do it
manually; these workarounds also make the version check useless
at the same time.

Moreover, most other BPF tracing types enabling bpf_probe_read()-like
functionality have /not/ adapted this check, and in general these
days it is well understood anyway that all the tracing programs are
not stable with regards to future kernels as kernel internal data
structures are subject to change from release to release.

Back at last netconf we discussed [0] and agreed to remove this
check from bpf_prog_load() and instead document it here in the uapi
header that there is no such guarantee for stable API for these
programs.

  [0] http://vger.kernel.org/netconf2018_files/DanielBorkmann_netconf2018.pdf

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Quentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index e7d57e89f25fc7932c4af6a64f23203e93e30e65..1d324c2cbca2d277a187599e3137b3c4c3e55ab3 100644 (file)
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -133,6 +133,14 @@ enum bpf_map_type {
        BPF_MAP_TYPE_STACK,
 };
 
+/* Note that tracing related programs such as
+ * BPF_PROG_TYPE_{KPROBE,TRACEPOINT,PERF_EVENT,RAW_TRACEPOINT}
+ * are not subject to a stable API since kernel internal data
+ * structures can change from release to release and may
+ * therefore break existing tracing BPF programs. Tracing BPF
+ * programs correspond to /a/ specific kernel which is to be
+ * analyzed, and not /a/ specific kernel /and/ all future ones.
+ */
 enum bpf_prog_type {
        BPF_PROG_TYPE_UNSPEC,
        BPF_PROG_TYPE_SOCKET_FILTER,
@@ -343,7 +351,7 @@ union bpf_attr {
                __u32           log_level;      /* verbosity level of verifier */
                __u32           log_size;       /* size of user buffer */
                __aligned_u64   log_buf;        /* user supplied buffer */
-               __u32           kern_version;   /* checked when prog_type=kprobe */
+               __u32           kern_version;   /* not used */
                __u32           prog_flags;
                char            prog_name[BPF_OBJ_NAME_LEN];
                __u32           prog_ifindex;   /* ifindex of netdev to prep for */

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 6ae062f1cf20d631f9e10ad58104ca943e4e587e..5db31067d85ef8264f06541c492648e70e89e713 100644 (file)
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1473,11 +1473,6 @@ static int bpf_prog_load(union bpf_attr *attr, union bpf_attr __user *uattr)
 
        if (attr->insn_cnt == 0 || attr->insn_cnt > BPF_MAXINSNS)
                return -E2BIG;
-
-       if (type == BPF_PROG_TYPE_KPROBE &&
-           attr->kern_version != LINUX_VERSION_CODE)
-               return -EINVAL;
-
        if (type != BPF_PROG_TYPE_SOCKET_FILTER &&
            type != BPF_PROG_TYPE_CGROUP_SKB &&
            !capable(CAP_SYS_ADMIN))

diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index e7d57e89f25fc7932c4af6a64f23203e93e30e65..1d324c2cbca2d277a187599e3137b3c4c3e55ab3 100644 (file)
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -133,6 +133,14 @@ enum bpf_map_type {
        BPF_MAP_TYPE_STACK,
 };
 
+/* Note that tracing related programs such as
+ * BPF_PROG_TYPE_{KPROBE,TRACEPOINT,PERF_EVENT,RAW_TRACEPOINT}
+ * are not subject to a stable API since kernel internal data
+ * structures can change from release to release and may
+ * therefore break existing tracing BPF programs. Tracing BPF
+ * programs correspond to /a/ specific kernel which is to be
+ * analyzed, and not /a/ specific kernel /and/ all future ones.
+ */
 enum bpf_prog_type {
        BPF_PROG_TYPE_UNSPEC,
        BPF_PROG_TYPE_SOCKET_FILTER,
@@ -343,7 +351,7 @@ union bpf_attr {
                __u32           log_level;      /* verbosity level of verifier */
                __u32           log_size;       /* size of user buffer */
                __aligned_u64   log_buf;        /* user supplied buffer */
-               __u32           kern_version;   /* checked when prog_type=kprobe */
+               __u32           kern_version;   /* not used */
                __u32           prog_flags;
                char            prog_name[BPF_OBJ_NAME_LEN];
                __u32           prog_ifindex;   /* ifindex of netdev to prep for */