* [PATCH bpf-next 0/4] Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE to libbpf
@ 2020-02-29 23:11 Andrii Nakryiko
2020-02-29 23:11 ` [PATCH bpf-next 1/4] bpftool: add header guards to generated vmlinux.h Andrii Nakryiko
` (4 more replies)
0 siblings, 5 replies; 6+ messages in thread
From: Andrii Nakryiko @ 2020-02-29 23:11 UTC (permalink / raw)
To: bpf, netdev, ast, daniel; +Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE helper macros from private
selftests helpers to public libbpf ones. These helpers are extremely helpful
for writing tracing BPF applications and have been requested to be exposed for
easy use (e.g., [0]).
As part of this move, fix up BPF_KRETPROBE to not allow for capturing input
arguments (as it's unreliable and they will be often clobbered). Also, add
vmlinux.h header guard to allow multi-time inclusion, if necessary; but also
to let PT_REGS_PARM do proper detection of struct pt_regs field names on x86
arch. See relevant patches for more details.
[0] https://github.com/iovisor/bcc/pull/2778#issue-381642907
Andrii Nakryiko (4):
bpftool: add header guards to generated vmlinux.h
libbpf: fix use of PT_REGS_PARM macros with vmlinux.h
selftests/bpf: fix BPF_KRETPROBE macro and use it in attach_probe test
libbpf: merge selftests' bpf_trace_helpers.h into libbpf's
bpf_tracing.h
tools/bpf/bpftool/btf.c | 5 +
tools/lib/bpf/bpf_tracing.h | 120 +++++++++++++++++-
tools/testing/selftests/bpf/bpf_tcp_helpers.h | 2 +-
.../testing/selftests/bpf/bpf_trace_helpers.h | 120 ------------------
tools/testing/selftests/bpf/progs/bpf_dctcp.c | 2 +-
.../testing/selftests/bpf/progs/fentry_test.c | 2 +-
.../selftests/bpf/progs/fexit_bpf2bpf.c | 2 +-
.../bpf/progs/fexit_bpf2bpf_simple.c | 2 +-
.../testing/selftests/bpf/progs/fexit_test.c | 2 +-
tools/testing/selftests/bpf/progs/kfree_skb.c | 2 +-
.../selftests/bpf/progs/test_attach_probe.c | 3 +-
.../selftests/bpf/progs/test_overhead.c | 7 +-
.../selftests/bpf/progs/test_perf_branches.c | 2 +-
.../selftests/bpf/progs/test_perf_buffer.c | 2 +-
.../selftests/bpf/progs/test_probe_user.c | 1 -
.../bpf/progs/test_trampoline_count.c | 3 +-
.../selftests/bpf/progs/test_xdp_bpf2bpf.c | 2 +-
17 files changed, 140 insertions(+), 139 deletions(-)
delete mode 100644 tools/testing/selftests/bpf/bpf_trace_helpers.h
--
2.17.1
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH bpf-next 1/4] bpftool: add header guards to generated vmlinux.h
2020-02-29 23:11 [PATCH bpf-next 0/4] Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE to libbpf Andrii Nakryiko
@ 2020-02-29 23:11 ` Andrii Nakryiko
2020-02-29 23:11 ` [PATCH bpf-next 2/4] libbpf: fix use of PT_REGS_PARM macros with vmlinux.h Andrii Nakryiko
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: Andrii Nakryiko @ 2020-02-29 23:11 UTC (permalink / raw)
To: bpf, netdev, ast, daniel; +Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
Add canonical #ifndef/#define/#endif guard for generated vmlinux.h header with
__VMLINUX_H__ symbol. __VMLINUX_H__ is also going to play double role of
identifying whether vmlinux.h is being used, versus, say, BCC or non-CO-RE
libbpf modes with dependency on kernel headers. This will make it possible to
write helper macro/functions, agnostic to exact BPF program set up.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
---
tools/bpf/bpftool/btf.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/tools/bpf/bpftool/btf.c b/tools/bpf/bpftool/btf.c
index b3745ed711ba..bcaf55b59498 100644
--- a/tools/bpf/bpftool/btf.c
+++ b/tools/bpf/bpftool/btf.c
@@ -389,6 +389,9 @@ static int dump_btf_c(const struct btf *btf,
if (IS_ERR(d))
return PTR_ERR(d);
+ printf("#ifndef __VMLINUX_H__\n");
+ printf("#define __VMLINUX_H__\n");
+ printf("\n");
printf("#ifndef BPF_NO_PRESERVE_ACCESS_INDEX\n");
printf("#pragma clang attribute push (__attribute__((preserve_access_index)), apply_to = record)\n");
printf("#endif\n\n");
@@ -412,6 +415,8 @@ static int dump_btf_c(const struct btf *btf,
printf("#ifndef BPF_NO_PRESERVE_ACCESS_INDEX\n");
printf("#pragma clang attribute pop\n");
printf("#endif\n");
+ printf("\n");
+ printf("#endif /* __VMLINUX_H__ */\n");
done:
btf_dump__free(d);
--
2.17.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH bpf-next 2/4] libbpf: fix use of PT_REGS_PARM macros with vmlinux.h
2020-02-29 23:11 [PATCH bpf-next 0/4] Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE to libbpf Andrii Nakryiko
2020-02-29 23:11 ` [PATCH bpf-next 1/4] bpftool: add header guards to generated vmlinux.h Andrii Nakryiko
@ 2020-02-29 23:11 ` Andrii Nakryiko
2020-02-29 23:11 ` [PATCH bpf-next 3/4] selftests/bpf: fix BPF_KRETPROBE macro and use it in attach_probe test Andrii Nakryiko
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: Andrii Nakryiko @ 2020-02-29 23:11 UTC (permalink / raw)
To: bpf, netdev, ast, daniel; +Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
Add detection of vmlinux.h to bpf_tracing.h header for PT_REGS macro.
Currently, BPF applications have to define __KERNEL__ symbol to use correct
definition of struct pt_regs on x86 arch. This is due to different field names
under internal kernel vs UAPI conditions. To make this more transparent for
users, detect vmlinux.h by checking __VMLINUX_H__ symbol.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
---
tools/lib/bpf/bpf_tracing.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/lib/bpf/bpf_tracing.h b/tools/lib/bpf/bpf_tracing.h
index b0dafe8b4ebc..8376f22b0e36 100644
--- a/tools/lib/bpf/bpf_tracing.h
+++ b/tools/lib/bpf/bpf_tracing.h
@@ -49,7 +49,7 @@
#if defined(bpf_target_x86)
-#ifdef __KERNEL__
+#if defined(__KERNEL__) || defined(__VMLINUX_H__)
#define PT_REGS_PARM1(x) ((x)->di)
#define PT_REGS_PARM2(x) ((x)->si)
#define PT_REGS_PARM3(x) ((x)->dx)
--
2.17.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH bpf-next 3/4] selftests/bpf: fix BPF_KRETPROBE macro and use it in attach_probe test
2020-02-29 23:11 [PATCH bpf-next 0/4] Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE to libbpf Andrii Nakryiko
2020-02-29 23:11 ` [PATCH bpf-next 1/4] bpftool: add header guards to generated vmlinux.h Andrii Nakryiko
2020-02-29 23:11 ` [PATCH bpf-next 2/4] libbpf: fix use of PT_REGS_PARM macros with vmlinux.h Andrii Nakryiko
@ 2020-02-29 23:11 ` Andrii Nakryiko
2020-02-29 23:11 ` [PATCH bpf-next 4/4] libbpf: merge selftests' bpf_trace_helpers.h into libbpf's bpf_tracing.h Andrii Nakryiko
2020-03-03 0:30 ` [PATCH bpf-next 0/4] Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE to libbpf Alexei Starovoitov
4 siblings, 0 replies; 6+ messages in thread
From: Andrii Nakryiko @ 2020-02-29 23:11 UTC (permalink / raw)
To: bpf, netdev, ast, daniel; +Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
For kretprobes, there is no point in capturing input arguments from pt_regs,
as they are going to be, most probably, clobbered by the time probed kernel
function returns. So switch BPF_KRETPROBE to accept zero or one argument
(optional return result).
Fixes: ac065870d928 ("selftests/bpf: Add BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE macros")
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
---
tools/testing/selftests/bpf/bpf_trace_helpers.h | 13 +++++++------
.../testing/selftests/bpf/progs/test_attach_probe.c | 3 ++-
tools/testing/selftests/bpf/progs/test_overhead.c | 6 ++----
3 files changed, 11 insertions(+), 11 deletions(-)
diff --git a/tools/testing/selftests/bpf/bpf_trace_helpers.h b/tools/testing/selftests/bpf/bpf_trace_helpers.h
index c6f1354d93fb..83b8e02f5ee9 100644
--- a/tools/testing/selftests/bpf/bpf_trace_helpers.h
+++ b/tools/testing/selftests/bpf/bpf_trace_helpers.h
@@ -96,15 +96,16 @@ typeof(name(0)) name(struct pt_regs *ctx) \
static __always_inline typeof(name(0)) ____##name(struct pt_regs *ctx, ##args)
#define ___bpf_kretprobe_args0() ctx
-#define ___bpf_kretprobe_argsN(x, args...) \
- ___bpf_kprobe_args(args), (void *)PT_REGS_RET(ctx)
+#define ___bpf_kretprobe_args1(x) \
+ ___bpf_kretprobe_args0(), (void *)PT_REGS_RET(ctx)
#define ___bpf_kretprobe_args(args...) \
- ___bpf_apply(___bpf_kretprobe_args, ___bpf_empty(args))(args)
+ ___bpf_apply(___bpf_kretprobe_args, ___bpf_narg(args))(args)
/*
- * BPF_KRETPROBE is similar to BPF_KPROBE, except, in addition to listing all
- * input kprobe arguments, one last extra argument has to be specified, which
- * captures kprobe return value.
+ * BPF_KRETPROBE is similar to BPF_KPROBE, except, it only provides optional
+ * return value (in addition to `struct pt_regs *ctx`), but no input
+ * arguments, because they will be clobbered by the time probed function
+ * returns.
*/
#define BPF_KRETPROBE(name, args...) \
name(struct pt_regs *ctx); \
diff --git a/tools/testing/selftests/bpf/progs/test_attach_probe.c b/tools/testing/selftests/bpf/progs/test_attach_probe.c
index dd8fae6660ab..38ed8c3bf922 100644
--- a/tools/testing/selftests/bpf/progs/test_attach_probe.c
+++ b/tools/testing/selftests/bpf/progs/test_attach_probe.c
@@ -4,6 +4,7 @@
#include <linux/ptrace.h>
#include <linux/bpf.h>
#include <bpf/bpf_helpers.h>
+#include "bpf_trace_helpers.h"
int kprobe_res = 0;
int kretprobe_res = 0;
@@ -18,7 +19,7 @@ int handle_kprobe(struct pt_regs *ctx)
}
SEC("kretprobe/sys_nanosleep")
-int handle_kretprobe(struct pt_regs *ctx)
+int BPF_KRETPROBE(handle_kretprobe)
{
kretprobe_res = 2;
return 0;
diff --git a/tools/testing/selftests/bpf/progs/test_overhead.c b/tools/testing/selftests/bpf/progs/test_overhead.c
index bfe9fbcb9684..f43714c69cc8 100644
--- a/tools/testing/selftests/bpf/progs/test_overhead.c
+++ b/tools/testing/selftests/bpf/progs/test_overhead.c
@@ -17,11 +17,9 @@ int BPF_KPROBE(prog1, struct task_struct *tsk, const char *buf, bool exec)
}
SEC("kretprobe/__set_task_comm")
-int BPF_KRETPROBE(prog2,
- struct task_struct *tsk, const char *buf, bool exec,
- int ret)
+int BPF_KRETPROBE(prog2, int ret)
{
- return !PT_REGS_PARM1(ctx) && ret;
+ return ret;
}
SEC("raw_tp/task_rename")
--
2.17.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* [PATCH bpf-next 4/4] libbpf: merge selftests' bpf_trace_helpers.h into libbpf's bpf_tracing.h
2020-02-29 23:11 [PATCH bpf-next 0/4] Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE to libbpf Andrii Nakryiko
` (2 preceding siblings ...)
2020-02-29 23:11 ` [PATCH bpf-next 3/4] selftests/bpf: fix BPF_KRETPROBE macro and use it in attach_probe test Andrii Nakryiko
@ 2020-02-29 23:11 ` Andrii Nakryiko
2020-03-03 0:30 ` [PATCH bpf-next 0/4] Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE to libbpf Alexei Starovoitov
4 siblings, 0 replies; 6+ messages in thread
From: Andrii Nakryiko @ 2020-02-29 23:11 UTC (permalink / raw)
To: bpf, netdev, ast, daniel; +Cc: andrii.nakryiko, kernel-team, Andrii Nakryiko
Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE macro into libbpf's bpf_tracing.h
header to make it available for non-selftests users.
Signed-off-by: Andrii Nakryiko <andriin@fb.com>
---
tools/lib/bpf/bpf_tracing.h | 118 +++++++++++++++++
tools/testing/selftests/bpf/bpf_tcp_helpers.h | 2 +-
.../testing/selftests/bpf/bpf_trace_helpers.h | 121 ------------------
tools/testing/selftests/bpf/progs/bpf_dctcp.c | 2 +-
.../testing/selftests/bpf/progs/fentry_test.c | 2 +-
.../selftests/bpf/progs/fexit_bpf2bpf.c | 2 +-
.../bpf/progs/fexit_bpf2bpf_simple.c | 2 +-
.../testing/selftests/bpf/progs/fexit_test.c | 2 +-
tools/testing/selftests/bpf/progs/kfree_skb.c | 2 +-
.../selftests/bpf/progs/test_attach_probe.c | 2 +-
.../selftests/bpf/progs/test_overhead.c | 1 -
.../selftests/bpf/progs/test_perf_branches.c | 2 +-
.../selftests/bpf/progs/test_perf_buffer.c | 2 +-
.../selftests/bpf/progs/test_probe_user.c | 1 -
.../bpf/progs/test_trampoline_count.c | 3 +-
.../selftests/bpf/progs/test_xdp_bpf2bpf.c | 2 +-
16 files changed, 131 insertions(+), 135 deletions(-)
delete mode 100644 tools/testing/selftests/bpf/bpf_trace_helpers.h
diff --git a/tools/lib/bpf/bpf_tracing.h b/tools/lib/bpf/bpf_tracing.h
index 8376f22b0e36..379d03b211ea 100644
--- a/tools/lib/bpf/bpf_tracing.h
+++ b/tools/lib/bpf/bpf_tracing.h
@@ -192,4 +192,122 @@ struct pt_regs;
(void *)(PT_REGS_FP(ctx) + sizeof(ip))); })
#endif
+#define ___bpf_concat(a, b) a ## b
+#define ___bpf_apply(fn, n) ___bpf_concat(fn, n)
+#define ___bpf_nth(_, _1, _2, _3, _4, _5, _6, _7, _8, _9, _a, _b, _c, N, ...) N
+#define ___bpf_narg(...) \
+ ___bpf_nth(_, ##__VA_ARGS__, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0)
+#define ___bpf_empty(...) \
+ ___bpf_nth(_, ##__VA_ARGS__, N, N, N, N, N, N, N, N, N, N, 0)
+
+#define ___bpf_ctx_cast0() ctx
+#define ___bpf_ctx_cast1(x) ___bpf_ctx_cast0(), (void *)ctx[0]
+#define ___bpf_ctx_cast2(x, args...) ___bpf_ctx_cast1(args), (void *)ctx[1]
+#define ___bpf_ctx_cast3(x, args...) ___bpf_ctx_cast2(args), (void *)ctx[2]
+#define ___bpf_ctx_cast4(x, args...) ___bpf_ctx_cast3(args), (void *)ctx[3]
+#define ___bpf_ctx_cast5(x, args...) ___bpf_ctx_cast4(args), (void *)ctx[4]
+#define ___bpf_ctx_cast6(x, args...) ___bpf_ctx_cast5(args), (void *)ctx[5]
+#define ___bpf_ctx_cast7(x, args...) ___bpf_ctx_cast6(args), (void *)ctx[6]
+#define ___bpf_ctx_cast8(x, args...) ___bpf_ctx_cast7(args), (void *)ctx[7]
+#define ___bpf_ctx_cast9(x, args...) ___bpf_ctx_cast8(args), (void *)ctx[8]
+#define ___bpf_ctx_cast10(x, args...) ___bpf_ctx_cast9(args), (void *)ctx[9]
+#define ___bpf_ctx_cast11(x, args...) ___bpf_ctx_cast10(args), (void *)ctx[10]
+#define ___bpf_ctx_cast12(x, args...) ___bpf_ctx_cast11(args), (void *)ctx[11]
+#define ___bpf_ctx_cast(args...) \
+ ___bpf_apply(___bpf_ctx_cast, ___bpf_narg(args))(args)
+
+/*
+ * BPF_PROG is a convenience wrapper for generic tp_btf/fentry/fexit and
+ * similar kinds of BPF programs, that accept input arguments as a single
+ * pointer to untyped u64 array, where each u64 can actually be a typed
+ * pointer or integer of different size. Instead of requring user to write
+ * manual casts and work with array elements by index, BPF_PROG macro
+ * allows user to declare a list of named and typed input arguments in the
+ * same syntax as for normal C function. All the casting is hidden and
+ * performed transparently, while user code can just assume working with
+ * function arguments of specified type and name.
+ *
+ * Original raw context argument is preserved as well as 'ctx' argument.
+ * This is useful when using BPF helpers that expect original context
+ * as one of the parameters (e.g., for bpf_perf_event_output()).
+ */
+#define BPF_PROG(name, args...) \
+name(unsigned long long *ctx); \
+static __attribute__((always_inline)) typeof(name(0)) \
+____##name(unsigned long long *ctx, ##args); \
+typeof(name(0)) name(unsigned long long *ctx) \
+{ \
+ _Pragma("GCC diagnostic push") \
+ _Pragma("GCC diagnostic ignored \"-Wint-conversion\"") \
+ return ____##name(___bpf_ctx_cast(args)); \
+ _Pragma("GCC diagnostic pop") \
+} \
+static __attribute__((always_inline)) typeof(name(0)) \
+____##name(unsigned long long *ctx, ##args)
+
+struct pt_regs;
+
+#define ___bpf_kprobe_args0() ctx
+#define ___bpf_kprobe_args1(x) \
+ ___bpf_kprobe_args0(), (void *)PT_REGS_PARM1(ctx)
+#define ___bpf_kprobe_args2(x, args...) \
+ ___bpf_kprobe_args1(args), (void *)PT_REGS_PARM2(ctx)
+#define ___bpf_kprobe_args3(x, args...) \
+ ___bpf_kprobe_args2(args), (void *)PT_REGS_PARM3(ctx)
+#define ___bpf_kprobe_args4(x, args...) \
+ ___bpf_kprobe_args3(args), (void *)PT_REGS_PARM4(ctx)
+#define ___bpf_kprobe_args5(x, args...) \
+ ___bpf_kprobe_args4(args), (void *)PT_REGS_PARM5(ctx)
+#define ___bpf_kprobe_args(args...) \
+ ___bpf_apply(___bpf_kprobe_args, ___bpf_narg(args))(args)
+
+/*
+ * BPF_KPROBE serves the same purpose for kprobes as BPF_PROG for
+ * tp_btf/fentry/fexit BPF programs. It hides the underlying platform-specific
+ * low-level way of getting kprobe input arguments from struct pt_regs, and
+ * provides a familiar typed and named function arguments syntax and
+ * semantics of accessing kprobe input paremeters.
+ *
+ * Original struct pt_regs* context is preserved as 'ctx' argument. This might
+ * be necessary when using BPF helpers like bpf_perf_event_output().
+ */
+#define BPF_KPROBE(name, args...) \
+name(struct pt_regs *ctx); \
+static __attribute__((always_inline)) typeof(name(0)) \
+____##name(struct pt_regs *ctx, ##args); \
+typeof(name(0)) name(struct pt_regs *ctx) \
+{ \
+ _Pragma("GCC diagnostic push") \
+ _Pragma("GCC diagnostic ignored \"-Wint-conversion\"") \
+ return ____##name(___bpf_kprobe_args(args)); \
+ _Pragma("GCC diagnostic pop") \
+} \
+static __attribute__((always_inline)) typeof(name(0)) \
+____##name(struct pt_regs *ctx, ##args)
+
+#define ___bpf_kretprobe_args0() ctx
+#define ___bpf_kretprobe_args1(x) \
+ ___bpf_kretprobe_args0(), (void *)PT_REGS_RET(ctx)
+#define ___bpf_kretprobe_args(args...) \
+ ___bpf_apply(___bpf_kretprobe_args, ___bpf_narg(args))(args)
+
+/*
+ * BPF_KRETPROBE is similar to BPF_KPROBE, except, it only provides optional
+ * return value (in addition to `struct pt_regs *ctx`), but no input
+ * arguments, because they will be clobbered by the time probed function
+ * returns.
+ */
+#define BPF_KRETPROBE(name, args...) \
+name(struct pt_regs *ctx); \
+static __attribute__((always_inline)) typeof(name(0)) \
+____##name(struct pt_regs *ctx, ##args); \
+typeof(name(0)) name(struct pt_regs *ctx) \
+{ \
+ _Pragma("GCC diagnostic push") \
+ _Pragma("GCC diagnostic ignored \"-Wint-conversion\"") \
+ return ____##name(___bpf_kretprobe_args(args)); \
+ _Pragma("GCC diagnostic pop") \
+} \
+static __always_inline typeof(name(0)) ____##name(struct pt_regs *ctx, ##args)
+
#endif
diff --git a/tools/testing/selftests/bpf/bpf_tcp_helpers.h b/tools/testing/selftests/bpf/bpf_tcp_helpers.h
index 8f21965ffc6c..5bf2fe9b1efa 100644
--- a/tools/testing/selftests/bpf/bpf_tcp_helpers.h
+++ b/tools/testing/selftests/bpf/bpf_tcp_helpers.h
@@ -6,7 +6,7 @@
#include <linux/types.h>
#include <bpf/bpf_helpers.h>
#include <bpf/bpf_core_read.h>
-#include "bpf_trace_helpers.h"
+#include <bpf/bpf_tracing.h>
#define BPF_STRUCT_OPS(name, args...) \
SEC("struct_ops/"#name) \
diff --git a/tools/testing/selftests/bpf/bpf_trace_helpers.h b/tools/testing/selftests/bpf/bpf_trace_helpers.h
deleted file mode 100644
index 83b8e02f5ee9..000000000000
--- a/tools/testing/selftests/bpf/bpf_trace_helpers.h
+++ /dev/null
@@ -1,121 +0,0 @@
-/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
-#ifndef __BPF_TRACE_HELPERS_H
-#define __BPF_TRACE_HELPERS_H
-
-#include <bpf/bpf_helpers.h>
-
-#define ___bpf_concat(a, b) a ## b
-#define ___bpf_apply(fn, n) ___bpf_concat(fn, n)
-#define ___bpf_nth(_, _1, _2, _3, _4, _5, _6, _7, _8, _9, _a, _b, _c, N, ...) N
-#define ___bpf_narg(...) \
- ___bpf_nth(_, ##__VA_ARGS__, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0)
-#define ___bpf_empty(...) \
- ___bpf_nth(_, ##__VA_ARGS__, N, N, N, N, N, N, N, N, N, N, 0)
-
-#define ___bpf_ctx_cast0() ctx
-#define ___bpf_ctx_cast1(x) ___bpf_ctx_cast0(), (void *)ctx[0]
-#define ___bpf_ctx_cast2(x, args...) ___bpf_ctx_cast1(args), (void *)ctx[1]
-#define ___bpf_ctx_cast3(x, args...) ___bpf_ctx_cast2(args), (void *)ctx[2]
-#define ___bpf_ctx_cast4(x, args...) ___bpf_ctx_cast3(args), (void *)ctx[3]
-#define ___bpf_ctx_cast5(x, args...) ___bpf_ctx_cast4(args), (void *)ctx[4]
-#define ___bpf_ctx_cast6(x, args...) ___bpf_ctx_cast5(args), (void *)ctx[5]
-#define ___bpf_ctx_cast7(x, args...) ___bpf_ctx_cast6(args), (void *)ctx[6]
-#define ___bpf_ctx_cast8(x, args...) ___bpf_ctx_cast7(args), (void *)ctx[7]
-#define ___bpf_ctx_cast9(x, args...) ___bpf_ctx_cast8(args), (void *)ctx[8]
-#define ___bpf_ctx_cast10(x, args...) ___bpf_ctx_cast9(args), (void *)ctx[9]
-#define ___bpf_ctx_cast11(x, args...) ___bpf_ctx_cast10(args), (void *)ctx[10]
-#define ___bpf_ctx_cast12(x, args...) ___bpf_ctx_cast11(args), (void *)ctx[11]
-#define ___bpf_ctx_cast(args...) \
- ___bpf_apply(___bpf_ctx_cast, ___bpf_narg(args))(args)
-
-/*
- * BPF_PROG is a convenience wrapper for generic tp_btf/fentry/fexit and
- * similar kinds of BPF programs, that accept input arguments as a single
- * pointer to untyped u64 array, where each u64 can actually be a typed
- * pointer or integer of different size. Instead of requring user to write
- * manual casts and work with array elements by index, BPF_PROG macro
- * allows user to declare a list of named and typed input arguments in the
- * same syntax as for normal C function. All the casting is hidden and
- * performed transparently, while user code can just assume working with
- * function arguments of specified type and name.
- *
- * Original raw context argument is preserved as well as 'ctx' argument.
- * This is useful when using BPF helpers that expect original context
- * as one of the parameters (e.g., for bpf_perf_event_output()).
- */
-#define BPF_PROG(name, args...) \
-name(unsigned long long *ctx); \
-static __always_inline typeof(name(0)) \
-____##name(unsigned long long *ctx, ##args); \
-typeof(name(0)) name(unsigned long long *ctx) \
-{ \
- _Pragma("GCC diagnostic push") \
- _Pragma("GCC diagnostic ignored \"-Wint-conversion\"") \
- return ____##name(___bpf_ctx_cast(args)); \
- _Pragma("GCC diagnostic pop") \
-} \
-static __always_inline typeof(name(0)) \
-____##name(unsigned long long *ctx, ##args)
-
-struct pt_regs;
-
-#define ___bpf_kprobe_args0() ctx
-#define ___bpf_kprobe_args1(x) \
- ___bpf_kprobe_args0(), (void *)PT_REGS_PARM1(ctx)
-#define ___bpf_kprobe_args2(x, args...) \
- ___bpf_kprobe_args1(args), (void *)PT_REGS_PARM2(ctx)
-#define ___bpf_kprobe_args3(x, args...) \
- ___bpf_kprobe_args2(args), (void *)PT_REGS_PARM3(ctx)
-#define ___bpf_kprobe_args4(x, args...) \
- ___bpf_kprobe_args3(args), (void *)PT_REGS_PARM4(ctx)
-#define ___bpf_kprobe_args5(x, args...) \
- ___bpf_kprobe_args4(args), (void *)PT_REGS_PARM5(ctx)
-#define ___bpf_kprobe_args(args...) \
- ___bpf_apply(___bpf_kprobe_args, ___bpf_narg(args))(args)
-
-/*
- * BPF_KPROBE serves the same purpose for kprobes as BPF_PROG for
- * tp_btf/fentry/fexit BPF programs. It hides the underlying platform-specific
- * low-level way of getting kprobe input arguments from struct pt_regs, and
- * provides a familiar typed and named function arguments syntax and
- * semantics of accessing kprobe input paremeters.
- *
- * Original struct pt_regs* context is preserved as 'ctx' argument. This might
- * be necessary when using BPF helpers like bpf_perf_event_output().
- */
-#define BPF_KPROBE(name, args...) \
-name(struct pt_regs *ctx); \
-static __always_inline typeof(name(0)) ____##name(struct pt_regs *ctx, ##args);\
-typeof(name(0)) name(struct pt_regs *ctx) \
-{ \
- _Pragma("GCC diagnostic push") \
- _Pragma("GCC diagnostic ignored \"-Wint-conversion\"") \
- return ____##name(___bpf_kprobe_args(args)); \
- _Pragma("GCC diagnostic pop") \
-} \
-static __always_inline typeof(name(0)) ____##name(struct pt_regs *ctx, ##args)
-
-#define ___bpf_kretprobe_args0() ctx
-#define ___bpf_kretprobe_args1(x) \
- ___bpf_kretprobe_args0(), (void *)PT_REGS_RET(ctx)
-#define ___bpf_kretprobe_args(args...) \
- ___bpf_apply(___bpf_kretprobe_args, ___bpf_narg(args))(args)
-
-/*
- * BPF_KRETPROBE is similar to BPF_KPROBE, except, it only provides optional
- * return value (in addition to `struct pt_regs *ctx`), but no input
- * arguments, because they will be clobbered by the time probed function
- * returns.
- */
-#define BPF_KRETPROBE(name, args...) \
-name(struct pt_regs *ctx); \
-static __always_inline typeof(name(0)) ____##name(struct pt_regs *ctx, ##args);\
-typeof(name(0)) name(struct pt_regs *ctx) \
-{ \
- _Pragma("GCC diagnostic push") \
- _Pragma("GCC diagnostic ignored \"-Wint-conversion\"") \
- return ____##name(___bpf_kretprobe_args(args)); \
- _Pragma("GCC diagnostic pop") \
-} \
-static __always_inline typeof(name(0)) ____##name(struct pt_regs *ctx, ##args)
-#endif
diff --git a/tools/testing/selftests/bpf/progs/bpf_dctcp.c b/tools/testing/selftests/bpf/progs/bpf_dctcp.c
index b631fb5032d2..127ea762a062 100644
--- a/tools/testing/selftests/bpf/progs/bpf_dctcp.c
+++ b/tools/testing/selftests/bpf/progs/bpf_dctcp.c
@@ -9,7 +9,7 @@
#include <linux/bpf.h>
#include <linux/types.h>
#include <bpf/bpf_helpers.h>
-#include "bpf_trace_helpers.h"
+#include <bpf/bpf_tracing.h>
#include "bpf_tcp_helpers.h"
char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/fentry_test.c b/tools/testing/selftests/bpf/progs/fentry_test.c
index 38d3a82144ca..9365b686f84b 100644
--- a/tools/testing/selftests/bpf/progs/fentry_test.c
+++ b/tools/testing/selftests/bpf/progs/fentry_test.c
@@ -2,7 +2,7 @@
/* Copyright (c) 2019 Facebook */
#include <linux/bpf.h>
#include <bpf/bpf_helpers.h>
-#include "bpf_trace_helpers.h"
+#include <bpf/bpf_tracing.h>
char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
index c329fccf9842..98e1efe14549 100644
--- a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
+++ b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf.c
@@ -5,7 +5,7 @@
#include <linux/bpf.h>
#include <bpf/bpf_helpers.h>
#include <bpf/bpf_endian.h>
-#include "bpf_trace_helpers.h"
+#include <bpf/bpf_tracing.h>
struct sk_buff {
unsigned int len;
diff --git a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf_simple.c b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf_simple.c
index 92f3fa47cf40..85c0b516d6ee 100644
--- a/tools/testing/selftests/bpf/progs/fexit_bpf2bpf_simple.c
+++ b/tools/testing/selftests/bpf/progs/fexit_bpf2bpf_simple.c
@@ -2,7 +2,7 @@
/* Copyright (c) 2019 Facebook */
#include <linux/bpf.h>
#include <bpf/bpf_helpers.h>
-#include "bpf_trace_helpers.h"
+#include <bpf/bpf_tracing.h>
struct sk_buff {
unsigned int len;
diff --git a/tools/testing/selftests/bpf/progs/fexit_test.c b/tools/testing/selftests/bpf/progs/fexit_test.c
index 348109b9ea07..bd1e17d8024c 100644
--- a/tools/testing/selftests/bpf/progs/fexit_test.c
+++ b/tools/testing/selftests/bpf/progs/fexit_test.c
@@ -2,7 +2,7 @@
/* Copyright (c) 2019 Facebook */
#include <linux/bpf.h>
#include <bpf/bpf_helpers.h>
-#include "bpf_trace_helpers.h"
+#include <bpf/bpf_tracing.h>
char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/bpf/progs/kfree_skb.c b/tools/testing/selftests/bpf/progs/kfree_skb.c
index 8f48a909f079..a46a264ce24e 100644
--- a/tools/testing/selftests/bpf/progs/kfree_skb.c
+++ b/tools/testing/selftests/bpf/progs/kfree_skb.c
@@ -4,7 +4,7 @@
#include <stdbool.h>
#include <bpf/bpf_helpers.h>
#include <bpf/bpf_endian.h>
-#include "bpf_trace_helpers.h"
+#include <bpf/bpf_tracing.h>
char _license[] SEC("license") = "GPL";
struct {
diff --git a/tools/testing/selftests/bpf/progs/test_attach_probe.c b/tools/testing/selftests/bpf/progs/test_attach_probe.c
index 38ed8c3bf922..8056a4c6d918 100644
--- a/tools/testing/selftests/bpf/progs/test_attach_probe.c
+++ b/tools/testing/selftests/bpf/progs/test_attach_probe.c
@@ -4,7 +4,7 @@
#include <linux/ptrace.h>
#include <linux/bpf.h>
#include <bpf/bpf_helpers.h>
-#include "bpf_trace_helpers.h"
+#include <bpf/bpf_tracing.h>
int kprobe_res = 0;
int kretprobe_res = 0;
diff --git a/tools/testing/selftests/bpf/progs/test_overhead.c b/tools/testing/selftests/bpf/progs/test_overhead.c
index f43714c69cc8..56a50b25cd33 100644
--- a/tools/testing/selftests/bpf/progs/test_overhead.c
+++ b/tools/testing/selftests/bpf/progs/test_overhead.c
@@ -6,7 +6,6 @@
#include <linux/ptrace.h>
#include <bpf/bpf_helpers.h>
#include <bpf/bpf_tracing.h>
-#include "bpf_trace_helpers.h"
struct task_struct;
diff --git a/tools/testing/selftests/bpf/progs/test_perf_branches.c b/tools/testing/selftests/bpf/progs/test_perf_branches.c
index 0f7e27d97567..a1ccc831c882 100644
--- a/tools/testing/selftests/bpf/progs/test_perf_branches.c
+++ b/tools/testing/selftests/bpf/progs/test_perf_branches.c
@@ -5,7 +5,7 @@
#include <linux/ptrace.h>
#include <linux/bpf.h>
#include <bpf/bpf_helpers.h>
-#include "bpf_trace_helpers.h"
+#include <bpf/bpf_tracing.h>
int valid = 0;
int required_size_out = 0;
diff --git a/tools/testing/selftests/bpf/progs/test_perf_buffer.c b/tools/testing/selftests/bpf/progs/test_perf_buffer.c
index ebfcc9f50c35..ad59c4c9aba8 100644
--- a/tools/testing/selftests/bpf/progs/test_perf_buffer.c
+++ b/tools/testing/selftests/bpf/progs/test_perf_buffer.c
@@ -4,7 +4,7 @@
#include <linux/ptrace.h>
#include <linux/bpf.h>
#include <bpf/bpf_helpers.h>
-#include "bpf_trace_helpers.h"
+#include <bpf/bpf_tracing.h>
struct {
__uint(type, BPF_MAP_TYPE_PERF_EVENT_ARRAY);
diff --git a/tools/testing/selftests/bpf/progs/test_probe_user.c b/tools/testing/selftests/bpf/progs/test_probe_user.c
index d556b1572cc6..89b3532ccc75 100644
--- a/tools/testing/selftests/bpf/progs/test_probe_user.c
+++ b/tools/testing/selftests/bpf/progs/test_probe_user.c
@@ -7,7 +7,6 @@
#include <bpf/bpf_helpers.h>
#include <bpf/bpf_tracing.h>
-#include "bpf_trace_helpers.h"
static struct sockaddr_in old;
diff --git a/tools/testing/selftests/bpf/progs/test_trampoline_count.c b/tools/testing/selftests/bpf/progs/test_trampoline_count.c
index e51e6e3a81c2..f030e469d05b 100644
--- a/tools/testing/selftests/bpf/progs/test_trampoline_count.c
+++ b/tools/testing/selftests/bpf/progs/test_trampoline_count.c
@@ -2,7 +2,8 @@
#include <stdbool.h>
#include <stddef.h>
#include <linux/bpf.h>
-#include "bpf_trace_helpers.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
struct task_struct;
diff --git a/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c b/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c
index b840fc9e3ed5..42dd2fedd588 100644
--- a/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c
+++ b/tools/testing/selftests/bpf/progs/test_xdp_bpf2bpf.c
@@ -1,7 +1,7 @@
// SPDX-License-Identifier: GPL-2.0
#include <linux/bpf.h>
+#include <bpf/bpf_tracing.h>
#include <bpf/bpf_helpers.h>
-#include "bpf_trace_helpers.h"
struct net_device {
/* Structure does not need to contain all entries,
--
2.17.1
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH bpf-next 0/4] Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE to libbpf
2020-02-29 23:11 [PATCH bpf-next 0/4] Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE to libbpf Andrii Nakryiko
` (3 preceding siblings ...)
2020-02-29 23:11 ` [PATCH bpf-next 4/4] libbpf: merge selftests' bpf_trace_helpers.h into libbpf's bpf_tracing.h Andrii Nakryiko
@ 2020-03-03 0:30 ` Alexei Starovoitov
4 siblings, 0 replies; 6+ messages in thread
From: Alexei Starovoitov @ 2020-03-03 0:30 UTC (permalink / raw)
To: Andrii Nakryiko; +Cc: bpf, netdev, ast, daniel, andrii.nakryiko, kernel-team
On Sat, Feb 29, 2020 at 03:11:08PM -0800, Andrii Nakryiko wrote:
> Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE helper macros from private
> selftests helpers to public libbpf ones. These helpers are extremely helpful
> for writing tracing BPF applications and have been requested to be exposed for
> easy use (e.g., [0]).
>
> As part of this move, fix up BPF_KRETPROBE to not allow for capturing input
> arguments (as it's unreliable and they will be often clobbered). Also, add
> vmlinux.h header guard to allow multi-time inclusion, if necessary; but also
> to let PT_REGS_PARM do proper detection of struct pt_regs field names on x86
> arch. See relevant patches for more details.
>
> [0] https://github.com/iovisor/bcc/pull/2778#issue-381642907
Looks like folks started to copy paste this macro into different places.
So definitely makes sense to move it.
Applied. Thanks
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2020-03-03 0:30 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-02-29 23:11 [PATCH bpf-next 0/4] Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE to libbpf Andrii Nakryiko
2020-02-29 23:11 ` [PATCH bpf-next 1/4] bpftool: add header guards to generated vmlinux.h Andrii Nakryiko
2020-02-29 23:11 ` [PATCH bpf-next 2/4] libbpf: fix use of PT_REGS_PARM macros with vmlinux.h Andrii Nakryiko
2020-02-29 23:11 ` [PATCH bpf-next 3/4] selftests/bpf: fix BPF_KRETPROBE macro and use it in attach_probe test Andrii Nakryiko
2020-02-29 23:11 ` [PATCH bpf-next 4/4] libbpf: merge selftests' bpf_trace_helpers.h into libbpf's bpf_tracing.h Andrii Nakryiko
2020-03-03 0:30 ` [PATCH bpf-next 0/4] Move BPF_PROG, BPF_KPROBE, and BPF_KRETPROBE to libbpf Alexei Starovoitov
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).