Skip to contentSkip to conversation

selftests/bpf: trampoline_count: use dedicated bpf_testmod target

PATCH Superseded Sun Jian 5 participants 10 messages Mar 19

Subsystems (5)
Networking Kselftest BPF Arm STM32 +1 more subsystems

v1 of 4 Superseded Latest is v4 →

lore
Share & tools
lore

Sun Jian

author
#0 lore raw

PATCH 1/1 selftests/bpf: trampoline_count: use dedicated bpf_testmod target

trampoline_count fills all trampoline attachment slots for a single target function and verifies that one extra attach fails with -E2BIG.

It currently targets bpf_modify_return_test, which is also used by other selftests such as modify_return, get_func_ip_test, and get_func_args_test. When such tests run in parallel, they can contend for the same per-function trampoline quota and cause unexpected attach failures. This issue is currently masked by harness serialization.

Move trampoline_count to a dedicated bpf_testmod target and register it for fmod_ret attachment. This keeps the test semantics unchanged while isolating it from other selftests, so it no longer needs to run in serial mode. Remove the TODO comment as well.

Tested:
./test_progs -t trampoline_count -vv
./test_progs -t modify_return -vv
./test_progs -t get_func_ip_test -vv
./test_progs -t get_func_args_test -vv
./test_progs -j$(nproc) -t trampoline_count -vv
./test_progs -j$(nproc) -t

    trampoline_count,modify_return,get_func_ip_test,get_func_args_test,\
kprobe_multi_test -vv
  20 runs of:
    ./test_progs -j$(nproc) -t
    trampoline_count,modify_return,get_func_ip_test,get_func_args_test,\

kprobe_multi_test

---
 .../selftests/bpf/prog_tests/trampoline_count.c |  3 +--
 .../selftests/bpf/progs/test_trampoline_count.c |  6 +++---
 .../selftests/bpf/test_kmods/bpf_testmod.c      | 17 +++++++++++++++++
 3 files changed, 21 insertions(+), 5 deletions(-)
tools/testing/selftests/bpf/prog_tests/trampoline_count.c +1 -2
index 6cd7349d4a2b..b19358aed126 100644
@@ -30,8 +30,7 @@ static struct bpf_program *load_prog(char *file, char *name, struct inst *inst)
 	return prog;
 }
 
-/* TODO: use different target function to run in concurrent mode */
-void serial_test_trampoline_count(void)
+void test_trampoline_count(void)
 {
 	char *file = "test_trampoline_count.bpf.o";
 	char *const progs[] = { "fentry_test", "fmod_ret_test", "fexit_test" };
tools/testing/selftests/bpf/progs/test_trampoline_count.c +3 -3
index 7765720da7d5..14ad2f53cf33 100644
@@ -3,19 +3,19 @@
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
 
-SEC("fentry/bpf_modify_return_test")
+SEC("fentry/bpf_testmod_trampoline_count_test")
 int BPF_PROG(fentry_test, int a, int *b)
 {
 	return 0;
 }
 
-SEC("fmod_ret/bpf_modify_return_test")
+SEC("fmod_ret/bpf_testmod_trampoline_count_test")
 int BPF_PROG(fmod_ret_test, int a, int *b, int ret)
 {
 	return 0;
 }
 
-SEC("fexit/bpf_modify_return_test")
+SEC("fexit/bpf_testmod_trampoline_count_test")
 int BPF_PROG(fexit_test, int a, int *b, int ret)
 {
 	return 0;
tools/testing/selftests/bpf/test_kmods/bpf_testmod.c +17 -0
index e62c6b78657f..6eaceb4dfbec 100644
@@ -581,6 +581,12 @@ noinline int bpf_fentry_shadow_test(int a)
 }
 EXPORT_SYMBOL_GPL(bpf_fentry_shadow_test);
 
+noinline int bpf_testmod_trampoline_count_test(int a, int *b)
+{
+	*b += 1;
+	return a + *b;
+}
+
 __bpf_hook_end();
 
 static struct bin_attribute bin_attr_bpf_testmod_file __ro_after_init = {
@@ -1843,6 +1849,16 @@ struct bpf_struct_ops testmod_multi_st_ops = {
 
 extern int bpf_fentry_test1(int a);
 
+BTF_KFUNCS_START(bpf_testmod_trampoline_count_ids)
+BTF_ID_FLAGS(func, bpf_testmod_trampoline_count_test)
+BTF_KFUNCS_END(bpf_testmod_trampoline_count_ids)
+
+static const struct
+btf_kfunc_id_set bpf_testmod_trampoline_count_fmodret_set = {
+	.owner = THIS_MODULE,
+	.set = &bpf_testmod_trampoline_count_ids,
+};
+
 static int bpf_testmod_init(void)
 {
 	const struct btf_id_dtor_kfunc bpf_testmod_dtors[] = {
@@ -1859,6 +1875,7 @@ static int bpf_testmod_init(void)
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_TRACING, &bpf_testmod_kfunc_set);
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_SYSCALL, &bpf_testmod_kfunc_set);
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &bpf_testmod_kfunc_set);
+	ret = ret ?: register_btf_fmodret_id_set(&bpf_testmod_trampoline_count_fmodret_set);
 	ret = ret ?: register_bpf_struct_ops(&bpf_bpf_testmod_ops, bpf_testmod_ops);
 	ret = ret ?: register_bpf_struct_ops(&bpf_testmod_ops2, bpf_testmod_ops2);
 	ret = ret ?: register_bpf_struct_ops(&bpf_testmod_ops3, bpf_testmod_ops3);

base-commit: 8a30aeb0d1b4e4aaf7f7bae72f20f2ae75385ccb

Signature · 1 line
2.43.0
#1 lore raw
Quoted 3 lines · Sun Jian

trampoline_count fills all trampoline attachment slots for a single target function and verifies that one extra attach fails with -E2BIG.

selftests/bpf: trampoline_count:

don't invent new subjects.
Use
selftests/bpf: ...

and explain what it does.
'trampoline_count:' is not a generic prefix.

pw-bot: cr

author
#2 lore raw

PATCH v2 1/1 selftests/bpf: trampoline_count: use dedicated target function

In reply to Sun Jian

trampoline_count fills all trampoline attachment slots for a single target function and verifies that one extra attach fails with -E2BIG.

It currently targets bpf_modify_return_test, which is also used by modify_return and get_func_ip_test. When those tests run in parallel, they can contend for the same per-function trampoline quota and cause unexpected attach failures. This issue is currently masked by harness serialization.

Add a dedicated bpf_trampoline_count_test target and switch trampoline_count to use it. This keeps the test semantics unchanged while isolating it from other modify_return-based selftests, so it no longer needs to run in serial mode. Remove the TODO comment as well.

Tested:
./test_progs -t trampoline_count -vv ./test_progs -t modify_return -vv ./test_progs -t get_func_ip_test -vv ./test_progs -j$(nproc) -t trampoline_count -vv ./test_progs -j$(nproc) -t
trampoline_count,modify_return,get_func_ip_test -vv

---

v2:
- add Paul's Tested-by

 net/bpf/test_run.c                                        | 7 +++++++
 tools/testing/selftests/bpf/prog_tests/trampoline_count.c | 3 +--
 tools/testing/selftests/bpf/progs/test_trampoline_count.c | 6 +++---
 3 files changed, 11 insertions(+), 5 deletions(-)
net/bpf/test_run.c +7 -0
index 178c4738e63b..b41147a29f16 100644
@@ -580,6 +580,12 @@ __bpf_kfunc int bpf_modify_return_test2(int a, int *b, short c, int d,
 	return a + *b + c + d + (long)e + f + g;
 }
 
+__bpf_kfunc int bpf_trampoline_count_test(int a, int *b)
+{
+	*b += 1;
+	return a + *b;
+}
+
 __bpf_kfunc int bpf_modify_return_test_tp(int nonce)
 {
 	trace_bpf_trigger_tp(nonce);
@@ -635,6 +641,7 @@ BTF_KFUNCS_START(bpf_test_modify_return_ids)
 BTF_ID_FLAGS(func, bpf_modify_return_test)
 BTF_ID_FLAGS(func, bpf_modify_return_test2)
 BTF_ID_FLAGS(func, bpf_modify_return_test_tp)
+BTF_ID_FLAGS(func, bpf_trampoline_count_test)
 BTF_ID_FLAGS(func, bpf_fentry_test1, KF_SLEEPABLE)
 BTF_KFUNCS_END(bpf_test_modify_return_ids)
 
tools/testing/selftests/bpf/prog_tests/trampoline_count.c +1 -2
index 6cd7349d4a2b..b19358aed126 100644
@@ -30,8 +30,7 @@ static struct bpf_program *load_prog(char *file, char *name, struct inst *inst)
 	return prog;
 }
 
-/* TODO: use different target function to run in concurrent mode */
-void serial_test_trampoline_count(void)
+void test_trampoline_count(void)
 {
 	char *file = "test_trampoline_count.bpf.o";
 	char *const progs[] = { "fentry_test", "fmod_ret_test", "fexit_test" };
tools/testing/selftests/bpf/progs/test_trampoline_count.c +3 -3
index 7765720da7d5..911d282fd423 100644
@@ -3,19 +3,19 @@
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
 
-SEC("fentry/bpf_modify_return_test")
+SEC("fentry/bpf_trampoline_count_test")
 int BPF_PROG(fentry_test, int a, int *b)
 {
 	return 0;
 }
 
-SEC("fmod_ret/bpf_modify_return_test")
+SEC("fmod_ret/bpf_trampoline_count_test")
 int BPF_PROG(fmod_ret_test, int a, int *b, int ret)
 {
 	return 0;
 }
 
-SEC("fexit/bpf_modify_return_test")
+SEC("fexit/bpf_trampoline_count_test")
 int BPF_PROG(fexit_test, int a, int *b, int ret)
 {
 	return 0;

base-commit: 267594792a71018788af69e836c52e34bb8054af

Signature · 1 line
2.43.0
#3 lore raw
In reply to Sun Jian
Alexei Starovoitov asked for the subject to not use "trampoline_count:" as a prefix, since it is not a standard subsystem prefix. The preferred form would be "selftests/bpf:" followed by a…
Quoted 1 line

selftests/bpf: trampoline_count: use dedicated bpf_testmod target

Alexei Starovoitov asked for the subject to not use "trampoline_count:" as a prefix, since it is not a standard subsystem prefix. The preferred form would be "selftests/bpf:" followed by a description of what the patch does.

https://lore.kernel.org/bpf/CAADnVQJ9HZTHKCzwmqUwhePbCQ5qPQrSd=yejUS17eHbMm0fuQ@mail.gmail.com/

---
AI reviewed your patch. Please fix the bug or email reply why it's not a bug.
See: https://github.com/kernel-patches/vmtest/blob/master/ci/claude/README.md

CI run summary: https://github.com/kernel-patches/bpf/actions/runs/23278379564

author
#4 lore raw

PATCH v2 1/1 selftests/bpf: move trampoline_count to dedicated bpf_testmod target

In reply to Sun Jian

trampoline_count fills all trampoline attachment slots for a single target function and verifies that one extra attach fails with -E2BIG.

It currently targets bpf_modify_return_test, which is also used by other selftests such as modify_return, get_func_ip_test, and get_func_args_test. When such tests run in parallel, they can contend for the same per-function trampoline quota and cause unexpected attach failures. This issue is currently masked by harness serialization.

Move trampoline_count to a dedicated bpf_testmod target and register it for fmod_ret attachment. This keeps the test semantics unchanged while isolating it from other selftests, so it no longer needs to run in serial mode. Remove the TODO comment as well.

Tested:
./test_progs -t trampoline_count -vv
./test_progs -t modify_return -vv
./test_progs -t get_func_ip_test -vv
./test_progs -t get_func_args_test -vv
./test_progs -j$(nproc) -t trampoline_count -vv
./test_progs -j$(nproc) -t

    trampoline_count,modify_return,get_func_ip_test,get_func_args_test,\
kprobe_multi_test -vv
  20 runs of:
    ./test_progs -j$(nproc) -t
    trampoline_count,modify_return,get_func_ip_test,get_func_args_test,\

kprobe_multi_test

---

v2:
- rewrite the subject to describe the change
- resend with the correct patch content
- resend as a reply to v1

 .../selftests/bpf/prog_tests/trampoline_count.c |  3 +--
 .../selftests/bpf/progs/test_trampoline_count.c |  6 +++---
 .../selftests/bpf/test_kmods/bpf_testmod.c      | 17 +++++++++++++++++
 3 files changed, 21 insertions(+), 5 deletions(-)
tools/testing/selftests/bpf/prog_tests/trampoline_count.c +1 -2
index 6cd7349d4a2b..b19358aed126 100644
@@ -30,8 +30,7 @@ static struct bpf_program *load_prog(char *file, char *name, struct inst *inst)
 	return prog;
 }
 
-/* TODO: use different target function to run in concurrent mode */
-void serial_test_trampoline_count(void)
+void test_trampoline_count(void)
 {
 	char *file = "test_trampoline_count.bpf.o";
 	char *const progs[] = { "fentry_test", "fmod_ret_test", "fexit_test" };
tools/testing/selftests/bpf/progs/test_trampoline_count.c +3 -3
index 7765720da7d5..14ad2f53cf33 100644
@@ -3,19 +3,19 @@
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
 
-SEC("fentry/bpf_modify_return_test")
+SEC("fentry/bpf_testmod_trampoline_count_test")
 int BPF_PROG(fentry_test, int a, int *b)
 {
 	return 0;
 }
 
-SEC("fmod_ret/bpf_modify_return_test")
+SEC("fmod_ret/bpf_testmod_trampoline_count_test")
 int BPF_PROG(fmod_ret_test, int a, int *b, int ret)
 {
 	return 0;
 }
 
-SEC("fexit/bpf_modify_return_test")
+SEC("fexit/bpf_testmod_trampoline_count_test")
 int BPF_PROG(fexit_test, int a, int *b, int ret)
 {
 	return 0;
tools/testing/selftests/bpf/test_kmods/bpf_testmod.c +17 -0
index e62c6b78657f..6eaceb4dfbec 100644
@@ -581,6 +581,12 @@ noinline int bpf_fentry_shadow_test(int a)
 }
 EXPORT_SYMBOL_GPL(bpf_fentry_shadow_test);
 
+noinline int bpf_testmod_trampoline_count_test(int a, int *b)
+{
+	*b += 1;
+	return a + *b;
+}
+
 __bpf_hook_end();
 
 static struct bin_attribute bin_attr_bpf_testmod_file __ro_after_init = {
@@ -1843,6 +1849,16 @@ struct bpf_struct_ops testmod_multi_st_ops = {
 
 extern int bpf_fentry_test1(int a);
 
+BTF_KFUNCS_START(bpf_testmod_trampoline_count_ids)
+BTF_ID_FLAGS(func, bpf_testmod_trampoline_count_test)
+BTF_KFUNCS_END(bpf_testmod_trampoline_count_ids)
+
+static const struct
+btf_kfunc_id_set bpf_testmod_trampoline_count_fmodret_set = {
+	.owner = THIS_MODULE,
+	.set = &bpf_testmod_trampoline_count_ids,
+};
+
 static int bpf_testmod_init(void)
 {
 	const struct btf_id_dtor_kfunc bpf_testmod_dtors[] = {
@@ -1859,6 +1875,7 @@ static int bpf_testmod_init(void)
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_TRACING, &bpf_testmod_kfunc_set);
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_SYSCALL, &bpf_testmod_kfunc_set);
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &bpf_testmod_kfunc_set);
+	ret = ret ?: register_btf_fmodret_id_set(&bpf_testmod_trampoline_count_fmodret_set);
 	ret = ret ?: register_bpf_struct_ops(&bpf_bpf_testmod_ops, bpf_testmod_ops);
 	ret = ret ?: register_bpf_struct_ops(&bpf_testmod_ops2, bpf_testmod_ops2);
 	ret = ret ?: register_bpf_struct_ops(&bpf_testmod_ops3, bpf_testmod_ops3);

base-commit: 8a30aeb0d1b4e4aaf7f7bae72f20f2ae75385ccb

Signature · 1 line
2.43.0
#5 lore raw

Re: [PATCH v2] selftests/bpf: move trampoline_count to dedicated bpf_testmod target

Quoted 22 lines · Sun Jian

trampoline_count fills all trampoline attachment slots for a single target function and verifies that one extra attach fails with -E2BIG.

It currently targets bpf_modify_return_test, which is also used by other selftests such as modify_return, get_func_ip_test, and get_func_args_test. When such tests run in parallel, they can contend for the same per-function trampoline quota and cause unexpected attach failures. This issue is currently masked by harness serialization.

Move trampoline_count to a dedicated bpf_testmod target and register it for fmod_ret attachment. This keeps the test semantics unchanged while isolating it from other selftests, so it no longer needs to run in serial mode. Remove the TODO comment as well.

Tested:
./test_progs -t trampoline_count -vv
./test_progs -t modify_return -vv
./test_progs -t get_func_ip_test -vv
./test_progs -t get_func_args_test -vv
./test_progs -j$(nproc) -t trampoline_count -vv
./test_progs -j$(nproc) -t

Quoted 8 lines
    trampoline_count,modify_return,get_func_ip_test,get_func_args_test,\
kprobe_multi_test -vv
  20 runs of:
    ./test_progs -j$(nproc) -t
    trampoline_count,modify_return,get_func_ip_test,get_func_args_test,\

kprobe_multi_test

if you want to land these patches please participate in code reviews:
https://lore.kernel.org/bpf/CAADnVQ+TKKptnNB25V3=bcdybh5G6c2DyW2sYtXvyRaVnPN8MA@mail.gmail.com/

#7 lore raw

Re: [PATCH v2] selftests/bpf: move trampoline_count to dedicated bpf_testmod target

In reply to Sun Jian
Quoted 70 lines · Sun Jian

trampoline_count fills all trampoline attachment slots for a single target function and verifies that one extra attach fails with -E2BIG.

It currently targets bpf_modify_return_test, which is also used by other selftests such as modify_return, get_func_ip_test, and get_func_args_test. When such tests run in parallel, they can contend for the same per-function trampoline quota and cause unexpected attach failures. This issue is currently masked by harness serialization.

Move trampoline_count to a dedicated bpf_testmod target and register it for fmod_ret attachment. This keeps the test semantics unchanged while isolating it from other selftests, so it no longer needs to run in serial mode. Remove the TODO comment as well.

Tested:
./test_progs -t trampoline_count -vv
./test_progs -t modify_return -vv
./test_progs -t get_func_ip_test -vv
./test_progs -t get_func_args_test -vv
./test_progs -j$(nproc) -t trampoline_count -vv
./test_progs -j$(nproc) -t

    trampoline_count,modify_return,get_func_ip_test,get_func_args_test,\
kprobe_multi_test -vv
  20 runs of:
    ./test_progs -j$(nproc) -t
    trampoline_count,modify_return,get_func_ip_test,get_func_args_test,\

kprobe_multi_test

---

v2:
- rewrite the subject to describe the change
- resend with the correct patch content
- resend as a reply to v1

 .../selftests/bpf/prog_tests/trampoline_count.c |  3 +--
 .../selftests/bpf/progs/test_trampoline_count.c |  6 +++---
 .../selftests/bpf/test_kmods/bpf_testmod.c      | 17 +++++++++++++++++
 3 files changed, 21 insertions(+), 5 deletions(-)
tools/testing/selftests/bpf/prog_tests/trampoline_count.c +1 -2
index 6cd7349d4a2b..b19358aed126 100644
@@ -30,8 +30,7 @@ static struct bpf_program *load_prog(char *file, char *name, struct inst *inst)
 	return prog;
 }
 
-/* TODO: use different target function to run in concurrent mode */
-void serial_test_trampoline_count(void)
+void test_trampoline_count(void)
 {
 	char *file = "test_trampoline_count.bpf.o";
 	char *const progs[] = { "fentry_test", "fmod_ret_test", "fexit_test" };
tools/testing/selftests/bpf/progs/test_trampoline_count.c +1 -1
index 7765720da7d5..14ad2f53cf33 100644
@@ -3,19 +3,19 @@
 #include <bpf/bpf_helpers.h>
 #include <bpf/bpf_tracing.h>
 
-SEC("fentry/bpf_modify_return_test")
+SEC("fentry/bpf_testmod_trampoline_count_test")
 int BPF_PROG(fentry_test, int a, int *b)
 {
 	return 0;
 }
 

tools/testing/selftests/bpf/progs/test_trampoline_count.c line 19

Quoted 10 lines
tools/testing/selftests/bpf/progs/test_trampoline_count.c +2 -2
-SEC("fmod_ret/bpf_modify_return_test")
+SEC("fmod_ret/bpf_testmod_trampoline_count_test")
 int BPF_PROG(fmod_ret_test, int a, int *b, int ret)
 {
 	return 0;
 }
 
-SEC("fexit/bpf_modify_return_test")
+SEC("fexit/bpf_testmod_trampoline_count_test")
 int BPF_PROG(fexit_test, int a, int *b, int ret)

last part of the trampoline_count test executes the probe through the kernel's test_run which executes bpf_modify_return_test

I don't think it changes much because the programs are doing nothing I guess it's there just to make sure the link failure did not screw up anything

we could call bpf_testmod_trampoline_count_test from bpf_testmod_test_read and have the test to call trigger_module_test_read

jirka

Quoted 49 lines
tools/testing/selftests/bpf/progs/test_trampoline_count.c +0 -0
 {
 	return 0;
tools/testing/selftests/bpf/test_kmods/bpf_testmod.c +17 -0
index e62c6b78657f..6eaceb4dfbec 100644
@@ -581,6 +581,12 @@ noinline int bpf_fentry_shadow_test(int a)
 }
 EXPORT_SYMBOL_GPL(bpf_fentry_shadow_test);
 
+noinline int bpf_testmod_trampoline_count_test(int a, int *b)
+{
+	*b += 1;
+	return a + *b;
+}
+
 __bpf_hook_end();
 
 static struct bin_attribute bin_attr_bpf_testmod_file __ro_after_init = {
@@ -1843,6 +1849,16 @@ struct bpf_struct_ops testmod_multi_st_ops = {
 
 extern int bpf_fentry_test1(int a);
 
+BTF_KFUNCS_START(bpf_testmod_trampoline_count_ids)
+BTF_ID_FLAGS(func, bpf_testmod_trampoline_count_test)
+BTF_KFUNCS_END(bpf_testmod_trampoline_count_ids)
+
+static const struct
+btf_kfunc_id_set bpf_testmod_trampoline_count_fmodret_set = {
+	.owner = THIS_MODULE,
+	.set = &bpf_testmod_trampoline_count_ids,
+};
+
 static int bpf_testmod_init(void)
 {
 	const struct btf_id_dtor_kfunc bpf_testmod_dtors[] = {
@@ -1859,6 +1875,7 @@ static int bpf_testmod_init(void)
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_TRACING, &bpf_testmod_kfunc_set);
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_SYSCALL, &bpf_testmod_kfunc_set);
 	ret = ret ?: register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &bpf_testmod_kfunc_set);
+	ret = ret ?: register_btf_fmodret_id_set(&bpf_testmod_trampoline_count_fmodret_set);
 	ret = ret ?: register_bpf_struct_ops(&bpf_bpf_testmod_ops, bpf_testmod_ops);
 	ret = ret ?: register_bpf_struct_ops(&bpf_testmod_ops2, bpf_testmod_ops2);
 	ret = ret ?: register_bpf_struct_ops(&bpf_testmod_ops3, bpf_testmod_ops3);

base-commit: 8a30aeb0d1b4e4aaf7f7bae72f20f2ae75385ccb

Signature · 1 line
2.43.0
#8 lore raw

Re: [PATCH v2] selftests/bpf: move trampoline_count to dedicated bpf_testmod target

In reply to Sun Jian
Quoted 26 lines · Sun Jian

trampoline_count fills all trampoline attachment slots for a single target function and verifies that one extra attach fails with -E2BIG.

It currently targets bpf_modify_return_test, which is also used by other selftests such as modify_return, get_func_ip_test, and get_func_args_test. When such tests run in parallel, they can contend for the same per-function trampoline quota and cause unexpected attach failures. This issue is currently masked by harness serialization.

Move trampoline_count to a dedicated bpf_testmod target and register it for fmod_ret attachment. This keeps the test semantics unchanged while isolating it from other selftests, so it no longer needs to run in serial mode. Remove the TODO comment as well.

Tested:
./test_progs -t trampoline_count -vv
./test_progs -t modify_return -vv
./test_progs -t get_func_ip_test -vv
./test_progs -t get_func_args_test -vv
./test_progs -j$(nproc) -t trampoline_count -vv
./test_progs -j$(nproc) -t

    trampoline_count,modify_return,get_func_ip_test,get_func_args_test,\
kprobe_multi_test -vv
  20 runs of:
    ./test_progs -j$(nproc) -t
    trampoline_count,modify_return,get_func_ip_test,get_func_args_test,\
Quoted 8 lines

kprobe_multi_test

---

v2:
- rewrite the subject to describe the change
- resend with the correct patch content

Please increment the version number when making these changes. Otherwise it can be quite confusing to everyone. I believe this is the third variant of your v2 on the mailing list.

Quoted 1 line

- resend as a reply to v1

That can also be a bit confusing as it will fold the email into existing conversations. You can instead link to the previous version from the changelogs.

[...]

Paul

author
#9 lore raw

Re: [PATCH v2] selftests/bpf: move trampoline_count to dedicated bpf_testmod target

Quoted 9 lines · Paul Chaignon
Quoted 4 lines

v2:
- rewrite the subject to describe the change
- resend with the correct patch content

Please increment the version number when making these changes. Otherwise it can be quite confusing to everyone. I believe this is the third variant of your v2 on the mailing list.

Quoted 5 lines
Quoted 1 line

- resend as a reply to v1

That can also be a bit confusing as it will fold the email into existing conversations. You can instead link to the previous version from the changelogs.

Hi Paul,
Thanks for the note.

Will bump the version for the next revision and send it as a new thread.

Regards,
Sun Jian