* [PATCH bpf-next v2 1/2] libbpf: implement bpf_usdt_arg_size BPF function
@ 2025-02-24 23:57 Ihor Solodrai
2025-02-24 23:57 ` [PATCH bpf-next v2 2/2] selftests/bpf: test bpf_usdt_arg_size() function Ihor Solodrai
` (2 more replies)
0 siblings, 3 replies; 5+ messages in thread
From: Ihor Solodrai @ 2025-02-24 23:57 UTC (permalink / raw)
To: bpf; +Cc: andrii, ast, daniel, eddyz87, mykolal, kernel-team
Information about USDT argument size is implicitly stored in
__bpf_usdt_arg_spec, but currently it's not accessbile to BPF programs
that use USDT.
Implement bpf_sdt_arg_size() that returns the size of an USDT argument
in bytes.
v1->v2:
* do not add __bpf_usdt_arg_spec() helper
v1: https://lore.kernel.org/bpf/20250220215904.3362709-1-ihor.solodrai@linux.dev/
Suggested-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Ihor Solodrai <ihor.solodrai@linux.dev>
---
tools/lib/bpf/usdt.bpf.h | 32 ++++++++++++++++++++++++++++++++
1 file changed, 32 insertions(+)
diff --git a/tools/lib/bpf/usdt.bpf.h b/tools/lib/bpf/usdt.bpf.h
index b811f754939f..2a7865c8e3fe 100644
--- a/tools/lib/bpf/usdt.bpf.h
+++ b/tools/lib/bpf/usdt.bpf.h
@@ -108,6 +108,38 @@ int bpf_usdt_arg_cnt(struct pt_regs *ctx)
return spec->arg_cnt;
}
+/* Returns the size in bytes of the #*arg_num* (zero-indexed) USDT argument.
+ * Returns negative error if argument is not found or arg_num is invalid.
+ */
+static __always_inline
+int bpf_usdt_arg_size(struct pt_regs *ctx, __u64 arg_num)
+{
+ struct __bpf_usdt_arg_spec *arg_spec;
+ struct __bpf_usdt_spec *spec;
+ int spec_id;
+
+ spec_id = __bpf_usdt_spec_id(ctx);
+ if (spec_id < 0)
+ return -ESRCH;
+
+ spec = bpf_map_lookup_elem(&__bpf_usdt_specs, &spec_id);
+ if (!spec)
+ return -ESRCH;
+
+ if (arg_num >= BPF_USDT_MAX_ARG_CNT)
+ return -ENOENT;
+ barrier_var(arg_num);
+ if (arg_num >= spec->arg_cnt)
+ return -ENOENT;
+
+ arg_spec = &spec->args[arg_num];
+
+ /* arg_spec->arg_bitshift = 64 - arg_sz * 8
+ * so: arg_sz = (64 - arg_spec->arg_bitshift) / 8
+ */
+ return (unsigned int)(64 - arg_spec->arg_bitshift) / 8;
+}
+
/* Fetch USDT argument #*arg_num* (zero-indexed) and put its value into *res.
* Returns 0 on success; negative error, otherwise.
* On error *res is guaranteed to be set to zero.
--
2.48.1
^ permalink raw reply related [flat|nested] 5+ messages in thread* [PATCH bpf-next v2 2/2] selftests/bpf: test bpf_usdt_arg_size() function 2025-02-24 23:57 [PATCH bpf-next v2 1/2] libbpf: implement bpf_usdt_arg_size BPF function Ihor Solodrai @ 2025-02-24 23:57 ` Ihor Solodrai 2025-02-26 14:36 ` Jiri Olsa 2025-02-26 14:34 ` [PATCH bpf-next v2 1/2] libbpf: implement bpf_usdt_arg_size BPF function Jiri Olsa 2025-02-26 18:20 ` patchwork-bot+netdevbpf 2 siblings, 1 reply; 5+ messages in thread From: Ihor Solodrai @ 2025-02-24 23:57 UTC (permalink / raw) To: bpf; +Cc: andrii, ast, daniel, eddyz87, mykolal, kernel-team Update usdt tests to also check for correct behavior of bpf_usdt_arg_size(). Signed-off-by: Ihor Solodrai <ihor.solodrai@linux.dev> --- tools/testing/selftests/bpf/prog_tests/usdt.c | 11 ++++++++++- tools/testing/selftests/bpf/progs/test_usdt.c | 14 ++++++++++++++ 2 files changed, 24 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/bpf/prog_tests/usdt.c b/tools/testing/selftests/bpf/prog_tests/usdt.c index 56ed1eb9b527..495d66414b57 100644 --- a/tools/testing/selftests/bpf/prog_tests/usdt.c +++ b/tools/testing/selftests/bpf/prog_tests/usdt.c @@ -45,7 +45,7 @@ static void subtest_basic_usdt(void) LIBBPF_OPTS(bpf_usdt_opts, opts); struct test_usdt *skel; struct test_usdt__bss *bss; - int err; + int err, i; skel = test_usdt__open_and_load(); if (!ASSERT_OK_PTR(skel, "skel_open")) @@ -75,6 +75,7 @@ static void subtest_basic_usdt(void) ASSERT_EQ(bss->usdt0_cookie, 0xcafedeadbeeffeed, "usdt0_cookie"); ASSERT_EQ(bss->usdt0_arg_cnt, 0, "usdt0_arg_cnt"); ASSERT_EQ(bss->usdt0_arg_ret, -ENOENT, "usdt0_arg_ret"); + ASSERT_EQ(bss->usdt0_arg_size, -ENOENT, "usdt0_arg_size"); /* auto-attached usdt3 gets default zero cookie value */ ASSERT_EQ(bss->usdt3_cookie, 0, "usdt3_cookie"); @@ -86,6 +87,9 @@ static void subtest_basic_usdt(void) ASSERT_EQ(bss->usdt3_args[0], 1, "usdt3_arg1"); ASSERT_EQ(bss->usdt3_args[1], 42, "usdt3_arg2"); ASSERT_EQ(bss->usdt3_args[2], (uintptr_t)&bla, "usdt3_arg3"); + ASSERT_EQ(bss->usdt3_arg_sizes[0], 4, "usdt3_arg1_size"); + ASSERT_EQ(bss->usdt3_arg_sizes[1], 8, "usdt3_arg2_size"); + ASSERT_EQ(bss->usdt3_arg_sizes[2], 8, "usdt3_arg3_size"); /* auto-attached usdt12 gets default zero cookie value */ ASSERT_EQ(bss->usdt12_cookie, 0, "usdt12_cookie"); @@ -104,6 +108,11 @@ static void subtest_basic_usdt(void) ASSERT_EQ(bss->usdt12_args[10], nums[idx], "usdt12_arg11"); ASSERT_EQ(bss->usdt12_args[11], t1.y, "usdt12_arg12"); + int usdt12_expected_arg_sizes[12] = { 4, 4, 8, 8, 4, 8, 8, 8, 4, 2, 2, 1 }; + + for (i = 0; i < 12; i++) + ASSERT_EQ(bss->usdt12_arg_sizes[i], usdt12_expected_arg_sizes[i], "usdt12_arg_size"); + /* trigger_func() is marked __always_inline, so USDT invocations will be * inlined in two different places, meaning that each USDT will have * at least 2 different places to be attached to. This verifies that diff --git a/tools/testing/selftests/bpf/progs/test_usdt.c b/tools/testing/selftests/bpf/progs/test_usdt.c index 505aab9a5234..096488f47fbc 100644 --- a/tools/testing/selftests/bpf/progs/test_usdt.c +++ b/tools/testing/selftests/bpf/progs/test_usdt.c @@ -11,6 +11,7 @@ int usdt0_called; u64 usdt0_cookie; int usdt0_arg_cnt; int usdt0_arg_ret; +int usdt0_arg_size; SEC("usdt") int usdt0(struct pt_regs *ctx) @@ -26,6 +27,7 @@ int usdt0(struct pt_regs *ctx) usdt0_arg_cnt = bpf_usdt_arg_cnt(ctx); /* should return -ENOENT for any arg_num */ usdt0_arg_ret = bpf_usdt_arg(ctx, bpf_get_prandom_u32(), &tmp); + usdt0_arg_size = bpf_usdt_arg_size(ctx, bpf_get_prandom_u32()); return 0; } @@ -34,6 +36,7 @@ u64 usdt3_cookie; int usdt3_arg_cnt; int usdt3_arg_rets[3]; u64 usdt3_args[3]; +int usdt3_arg_sizes[3]; SEC("usdt//proc/self/exe:test:usdt3") int usdt3(struct pt_regs *ctx) @@ -50,12 +53,15 @@ int usdt3(struct pt_regs *ctx) usdt3_arg_rets[0] = bpf_usdt_arg(ctx, 0, &tmp); usdt3_args[0] = (int)tmp; + usdt3_arg_sizes[0] = bpf_usdt_arg_size(ctx, 0); usdt3_arg_rets[1] = bpf_usdt_arg(ctx, 1, &tmp); usdt3_args[1] = (long)tmp; + usdt3_arg_sizes[1] = bpf_usdt_arg_size(ctx, 1); usdt3_arg_rets[2] = bpf_usdt_arg(ctx, 2, &tmp); usdt3_args[2] = (uintptr_t)tmp; + usdt3_arg_sizes[2] = bpf_usdt_arg_size(ctx, 2); return 0; } @@ -64,12 +70,15 @@ int usdt12_called; u64 usdt12_cookie; int usdt12_arg_cnt; u64 usdt12_args[12]; +int usdt12_arg_sizes[12]; SEC("usdt//proc/self/exe:test:usdt12") int BPF_USDT(usdt12, int a1, int a2, long a3, long a4, unsigned a5, long a6, __u64 a7, uintptr_t a8, int a9, short a10, short a11, signed char a12) { + int i; + if (my_pid != (bpf_get_current_pid_tgid() >> 32)) return 0; @@ -90,6 +99,11 @@ int BPF_USDT(usdt12, int a1, int a2, long a3, long a4, unsigned a5, usdt12_args[9] = a10; usdt12_args[10] = a11; usdt12_args[11] = a12; + + bpf_for(i, 0, 12) { + usdt12_arg_sizes[i] = bpf_usdt_arg_size(ctx, i); + } + return 0; } -- 2.48.1 ^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH bpf-next v2 2/2] selftests/bpf: test bpf_usdt_arg_size() function 2025-02-24 23:57 ` [PATCH bpf-next v2 2/2] selftests/bpf: test bpf_usdt_arg_size() function Ihor Solodrai @ 2025-02-26 14:36 ` Jiri Olsa 0 siblings, 0 replies; 5+ messages in thread From: Jiri Olsa @ 2025-02-26 14:36 UTC (permalink / raw) To: Ihor Solodrai; +Cc: bpf, andrii, ast, daniel, eddyz87, mykolal, kernel-team On Mon, Feb 24, 2025 at 03:57:56PM -0800, Ihor Solodrai wrote: > Update usdt tests to also check for correct behavior of > bpf_usdt_arg_size(). > > Signed-off-by: Ihor Solodrai <ihor.solodrai@linux.dev> Reviewed-by: Jiri Olsa <jolsa@kernel.org> jirka > --- > tools/testing/selftests/bpf/prog_tests/usdt.c | 11 ++++++++++- > tools/testing/selftests/bpf/progs/test_usdt.c | 14 ++++++++++++++ > 2 files changed, 24 insertions(+), 1 deletion(-) > > diff --git a/tools/testing/selftests/bpf/prog_tests/usdt.c b/tools/testing/selftests/bpf/prog_tests/usdt.c > index 56ed1eb9b527..495d66414b57 100644 > --- a/tools/testing/selftests/bpf/prog_tests/usdt.c > +++ b/tools/testing/selftests/bpf/prog_tests/usdt.c > @@ -45,7 +45,7 @@ static void subtest_basic_usdt(void) > LIBBPF_OPTS(bpf_usdt_opts, opts); > struct test_usdt *skel; > struct test_usdt__bss *bss; > - int err; > + int err, i; > > skel = test_usdt__open_and_load(); > if (!ASSERT_OK_PTR(skel, "skel_open")) > @@ -75,6 +75,7 @@ static void subtest_basic_usdt(void) > ASSERT_EQ(bss->usdt0_cookie, 0xcafedeadbeeffeed, "usdt0_cookie"); > ASSERT_EQ(bss->usdt0_arg_cnt, 0, "usdt0_arg_cnt"); > ASSERT_EQ(bss->usdt0_arg_ret, -ENOENT, "usdt0_arg_ret"); > + ASSERT_EQ(bss->usdt0_arg_size, -ENOENT, "usdt0_arg_size"); > > /* auto-attached usdt3 gets default zero cookie value */ > ASSERT_EQ(bss->usdt3_cookie, 0, "usdt3_cookie"); > @@ -86,6 +87,9 @@ static void subtest_basic_usdt(void) > ASSERT_EQ(bss->usdt3_args[0], 1, "usdt3_arg1"); > ASSERT_EQ(bss->usdt3_args[1], 42, "usdt3_arg2"); > ASSERT_EQ(bss->usdt3_args[2], (uintptr_t)&bla, "usdt3_arg3"); > + ASSERT_EQ(bss->usdt3_arg_sizes[0], 4, "usdt3_arg1_size"); > + ASSERT_EQ(bss->usdt3_arg_sizes[1], 8, "usdt3_arg2_size"); > + ASSERT_EQ(bss->usdt3_arg_sizes[2], 8, "usdt3_arg3_size"); > > /* auto-attached usdt12 gets default zero cookie value */ > ASSERT_EQ(bss->usdt12_cookie, 0, "usdt12_cookie"); > @@ -104,6 +108,11 @@ static void subtest_basic_usdt(void) > ASSERT_EQ(bss->usdt12_args[10], nums[idx], "usdt12_arg11"); > ASSERT_EQ(bss->usdt12_args[11], t1.y, "usdt12_arg12"); > > + int usdt12_expected_arg_sizes[12] = { 4, 4, 8, 8, 4, 8, 8, 8, 4, 2, 2, 1 }; > + > + for (i = 0; i < 12; i++) > + ASSERT_EQ(bss->usdt12_arg_sizes[i], usdt12_expected_arg_sizes[i], "usdt12_arg_size"); > + > /* trigger_func() is marked __always_inline, so USDT invocations will be > * inlined in two different places, meaning that each USDT will have > * at least 2 different places to be attached to. This verifies that > diff --git a/tools/testing/selftests/bpf/progs/test_usdt.c b/tools/testing/selftests/bpf/progs/test_usdt.c > index 505aab9a5234..096488f47fbc 100644 > --- a/tools/testing/selftests/bpf/progs/test_usdt.c > +++ b/tools/testing/selftests/bpf/progs/test_usdt.c > @@ -11,6 +11,7 @@ int usdt0_called; > u64 usdt0_cookie; > int usdt0_arg_cnt; > int usdt0_arg_ret; > +int usdt0_arg_size; > > SEC("usdt") > int usdt0(struct pt_regs *ctx) > @@ -26,6 +27,7 @@ int usdt0(struct pt_regs *ctx) > usdt0_arg_cnt = bpf_usdt_arg_cnt(ctx); > /* should return -ENOENT for any arg_num */ > usdt0_arg_ret = bpf_usdt_arg(ctx, bpf_get_prandom_u32(), &tmp); > + usdt0_arg_size = bpf_usdt_arg_size(ctx, bpf_get_prandom_u32()); > return 0; > } > > @@ -34,6 +36,7 @@ u64 usdt3_cookie; > int usdt3_arg_cnt; > int usdt3_arg_rets[3]; > u64 usdt3_args[3]; > +int usdt3_arg_sizes[3]; > > SEC("usdt//proc/self/exe:test:usdt3") > int usdt3(struct pt_regs *ctx) > @@ -50,12 +53,15 @@ int usdt3(struct pt_regs *ctx) > > usdt3_arg_rets[0] = bpf_usdt_arg(ctx, 0, &tmp); > usdt3_args[0] = (int)tmp; > + usdt3_arg_sizes[0] = bpf_usdt_arg_size(ctx, 0); > > usdt3_arg_rets[1] = bpf_usdt_arg(ctx, 1, &tmp); > usdt3_args[1] = (long)tmp; > + usdt3_arg_sizes[1] = bpf_usdt_arg_size(ctx, 1); > > usdt3_arg_rets[2] = bpf_usdt_arg(ctx, 2, &tmp); > usdt3_args[2] = (uintptr_t)tmp; > + usdt3_arg_sizes[2] = bpf_usdt_arg_size(ctx, 2); > > return 0; > } > @@ -64,12 +70,15 @@ int usdt12_called; > u64 usdt12_cookie; > int usdt12_arg_cnt; > u64 usdt12_args[12]; > +int usdt12_arg_sizes[12]; > > SEC("usdt//proc/self/exe:test:usdt12") > int BPF_USDT(usdt12, int a1, int a2, long a3, long a4, unsigned a5, > long a6, __u64 a7, uintptr_t a8, int a9, short a10, > short a11, signed char a12) > { > + int i; > + > if (my_pid != (bpf_get_current_pid_tgid() >> 32)) > return 0; > > @@ -90,6 +99,11 @@ int BPF_USDT(usdt12, int a1, int a2, long a3, long a4, unsigned a5, > usdt12_args[9] = a10; > usdt12_args[10] = a11; > usdt12_args[11] = a12; > + > + bpf_for(i, 0, 12) { > + usdt12_arg_sizes[i] = bpf_usdt_arg_size(ctx, i); > + } > + > return 0; > } > > -- > 2.48.1 > > ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH bpf-next v2 1/2] libbpf: implement bpf_usdt_arg_size BPF function 2025-02-24 23:57 [PATCH bpf-next v2 1/2] libbpf: implement bpf_usdt_arg_size BPF function Ihor Solodrai 2025-02-24 23:57 ` [PATCH bpf-next v2 2/2] selftests/bpf: test bpf_usdt_arg_size() function Ihor Solodrai @ 2025-02-26 14:34 ` Jiri Olsa 2025-02-26 18:20 ` patchwork-bot+netdevbpf 2 siblings, 0 replies; 5+ messages in thread From: Jiri Olsa @ 2025-02-26 14:34 UTC (permalink / raw) To: Ihor Solodrai; +Cc: bpf, andrii, ast, daniel, eddyz87, mykolal, kernel-team On Mon, Feb 24, 2025 at 03:57:55PM -0800, Ihor Solodrai wrote: > Information about USDT argument size is implicitly stored in > __bpf_usdt_arg_spec, but currently it's not accessbile to BPF programs > that use USDT. > > Implement bpf_sdt_arg_size() that returns the size of an USDT argument > in bytes. > > v1->v2: > * do not add __bpf_usdt_arg_spec() helper Reviewed-by: Jiri Olsa <jolsa@kernel.org> jirka > > v1: https://lore.kernel.org/bpf/20250220215904.3362709-1-ihor.solodrai@linux.dev/ > > Suggested-by: Andrii Nakryiko <andrii@kernel.org> > Signed-off-by: Ihor Solodrai <ihor.solodrai@linux.dev> > --- > tools/lib/bpf/usdt.bpf.h | 32 ++++++++++++++++++++++++++++++++ > 1 file changed, 32 insertions(+) > > diff --git a/tools/lib/bpf/usdt.bpf.h b/tools/lib/bpf/usdt.bpf.h > index b811f754939f..2a7865c8e3fe 100644 > --- a/tools/lib/bpf/usdt.bpf.h > +++ b/tools/lib/bpf/usdt.bpf.h > @@ -108,6 +108,38 @@ int bpf_usdt_arg_cnt(struct pt_regs *ctx) > return spec->arg_cnt; > } > > +/* Returns the size in bytes of the #*arg_num* (zero-indexed) USDT argument. > + * Returns negative error if argument is not found or arg_num is invalid. > + */ > +static __always_inline > +int bpf_usdt_arg_size(struct pt_regs *ctx, __u64 arg_num) > +{ > + struct __bpf_usdt_arg_spec *arg_spec; > + struct __bpf_usdt_spec *spec; > + int spec_id; > + > + spec_id = __bpf_usdt_spec_id(ctx); > + if (spec_id < 0) > + return -ESRCH; > + > + spec = bpf_map_lookup_elem(&__bpf_usdt_specs, &spec_id); > + if (!spec) > + return -ESRCH; > + > + if (arg_num >= BPF_USDT_MAX_ARG_CNT) > + return -ENOENT; > + barrier_var(arg_num); > + if (arg_num >= spec->arg_cnt) > + return -ENOENT; > + > + arg_spec = &spec->args[arg_num]; > + > + /* arg_spec->arg_bitshift = 64 - arg_sz * 8 > + * so: arg_sz = (64 - arg_spec->arg_bitshift) / 8 > + */ > + return (unsigned int)(64 - arg_spec->arg_bitshift) / 8; > +} > + > /* Fetch USDT argument #*arg_num* (zero-indexed) and put its value into *res. > * Returns 0 on success; negative error, otherwise. > * On error *res is guaranteed to be set to zero. > -- > 2.48.1 > > ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH bpf-next v2 1/2] libbpf: implement bpf_usdt_arg_size BPF function 2025-02-24 23:57 [PATCH bpf-next v2 1/2] libbpf: implement bpf_usdt_arg_size BPF function Ihor Solodrai 2025-02-24 23:57 ` [PATCH bpf-next v2 2/2] selftests/bpf: test bpf_usdt_arg_size() function Ihor Solodrai 2025-02-26 14:34 ` [PATCH bpf-next v2 1/2] libbpf: implement bpf_usdt_arg_size BPF function Jiri Olsa @ 2025-02-26 18:20 ` patchwork-bot+netdevbpf 2 siblings, 0 replies; 5+ messages in thread From: patchwork-bot+netdevbpf @ 2025-02-26 18:20 UTC (permalink / raw) To: Ihor Solodrai; +Cc: bpf, andrii, ast, daniel, eddyz87, mykolal, kernel-team Hello: This series was applied to bpf/bpf-next.git (master) by Andrii Nakryiko <andrii@kernel.org>: On Mon, 24 Feb 2025 15:57:55 -0800 you wrote: > Information about USDT argument size is implicitly stored in > __bpf_usdt_arg_spec, but currently it's not accessbile to BPF programs > that use USDT. > > Implement bpf_sdt_arg_size() that returns the size of an USDT argument > in bytes. > > [...] Here is the summary with links: - [bpf-next,v2,1/2] libbpf: implement bpf_usdt_arg_size BPF function https://git.kernel.org/bpf/bpf-next/c/b62dff14402a - [bpf-next,v2,2/2] selftests/bpf: test bpf_usdt_arg_size() function https://git.kernel.org/bpf/bpf-next/c/0ba0ef012eba You are awesome, thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html ^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2025-02-26 18:19 UTC | newest] Thread overview: 5+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2025-02-24 23:57 [PATCH bpf-next v2 1/2] libbpf: implement bpf_usdt_arg_size BPF function Ihor Solodrai 2025-02-24 23:57 ` [PATCH bpf-next v2 2/2] selftests/bpf: test bpf_usdt_arg_size() function Ihor Solodrai 2025-02-26 14:36 ` Jiri Olsa 2025-02-26 14:34 ` [PATCH bpf-next v2 1/2] libbpf: implement bpf_usdt_arg_size BPF function Jiri Olsa 2025-02-26 18:20 ` patchwork-bot+netdevbpf
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox