bpf: Prevent inlining of bpf_fentry_test7()
Subsystems (2)
Yonghong Song
reviewerPATCH 1/1 bpf: Prevent inlining of bpf_fentry_test7()
With latest clang18, I hit test_progs failures for the following test:
#13/2 bpf_cookie/multi_kprobe_link_api:FAIL #13/3 bpf_cookie/multi_kprobe_attach_api:FAIL #13 bpf_cookie:FAIL #75 fentry_fexit:FAIL #76/1 fentry_test/fentry:FAIL #76 fentry_test:FAIL #80/1 fexit_test/fexit:FAIL #80 fexit_test:FAIL #110/1 kprobe_multi_test/skel_api:FAIL #110/2 kprobe_multi_test/link_api_addrs:FAIL #110/3 kprobe_multi_test/link_api_syms:FAIL #110/4 kprobe_multi_test/attach_api_pattern:FAIL #110/5 kprobe_multi_test/attach_api_addrs:FAIL #110/6 kprobe_multi_test/attach_api_syms:FAIL #110 kprobe_multi_test:FAIL
For example, for #13/2, the error messages are
...
kprobe_multi_test_run:FAIL:kprobe_test7_result unexpected kprobe_test7_result: actual 0 != expected 1
...
kprobe_multi_test_run:FAIL:kretprobe_test7_result unexpected kretprobe_test7_result: actual 0 != expected 1
clang17 does not have this issue.
Further investigation shows that kernel func bpf_fentry_test7(), used in the above tests, is inlined by the compiler although it is marked as noinline.
int noinline bpf_fentry_test7(struct bpf_fentry_test_t *arg)
{
return (long)arg;
}It is known that for simple functions like the above (e.g. just returning a constant or an input argument), the clang compiler may still do inlining for a noinline function. Adding 'asm volatile ("")' in the beginning of the bpf_fentry_test7() can prevent inlining.
--- net/bpf/test_run.c | 1 + 1 file changed, 1 insertion(+)
net/bpf/test_run.c +1 -0
@@ -543,6 +543,7 @@ struct bpf_fentry_test_t { int noinline bpf_fentry_test7(struct bpf_fentry_test_t *arg) { + asmvolatile ("" );return (long )arg; }
Signature · 1 line
2.34.1
Quoted 34 lines · Yonghong Song
With latest clang18, I hit
test_progsfailures for the following test:#13/2 bpf_cookie/multi_kprobe_link_api:FAIL #13/3 bpf_cookie/multi_kprobe_attach_api:FAIL #13 bpf_cookie:FAIL #75 fentry_fexit:FAIL #76/1 fentry_test/fentry:FAIL #76 fentry_test:FAIL #80/1 fexit_test/fexit:FAIL #80 fexit_test:FAIL #110/1 kprobe_multi_test/skel_api:FAIL #110/2 kprobe_multi_test/link_api_addrs:FAIL #110/3 kprobe_multi_test/link_api_syms:FAIL #110/4 kprobe_multi_test/attach_api_pattern:FAIL #110/5 kprobe_multi_test/attach_api_addrs:FAIL #110/6 kprobe_multi_test/attach_api_syms:FAIL #110 kprobe_multi_test:FAILFor example, for #13/2, the error messages are
...
kprobe_multi_test_run:FAIL:kprobe_test7_result unexpectedkprobe_test7_result: actual 0 != expected 1
...
kprobe_multi_test_run:FAIL:kretprobe_test7_result unexpectedkretprobe_test7_result: actual 0 != expected 1clang17 does not have this issue.
Further investigation shows that kernel func
bpf_fentry_test7(), used in the above tests, is inlined by the compiler although it is marked as noinline.int noinline bpf_fentry_test7(struct bpf_fentry_test_t *arg) { return (long)arg; }
Quoted 4 lines
It is known that for simple functions like the above (e.g. just returning a constant or an input argument), the clang compiler may still do inlining for a noinline function. Adding 'asm volatile ("")' in the beginning of the
bpf_fentry_test7()can prevent inlining.
Can confirm, this patch fixes listed flaky tests for me. (when using LLVM main 651e644595b7 from 6 days ago).
Quoted 18 lines
Signed-off-by Yonghong Song <[email protected]>--- net/bpf/test_run.c | 1 + 1 file changed, 1 insertion(+)net/bpf/test_run.c +1
-0@@ -543,6 +543,7 @@ struct bpf_fentry_test_t { int noinline bpf_fentry_test7(struct bpf_fentry_test_t *arg) { + asmvolatile ("" );return (long )arg; }
Quoted 34 lines · Yonghong Song
With latest clang18, I hit
test_progsfailures for the following test:#13/2 bpf_cookie/multi_kprobe_link_api:FAIL #13/3 bpf_cookie/multi_kprobe_attach_api:FAIL #13 bpf_cookie:FAIL #75 fentry_fexit:FAIL #76/1 fentry_test/fentry:FAIL #76 fentry_test:FAIL #80/1 fexit_test/fexit:FAIL #80 fexit_test:FAIL #110/1 kprobe_multi_test/skel_api:FAIL #110/2 kprobe_multi_test/link_api_addrs:FAIL #110/3 kprobe_multi_test/link_api_syms:FAIL #110/4 kprobe_multi_test/attach_api_pattern:FAIL #110/5 kprobe_multi_test/attach_api_addrs:FAIL #110/6 kprobe_multi_test/attach_api_syms:FAIL #110 kprobe_multi_test:FAILFor example, for #13/2, the error messages are
...
kprobe_multi_test_run:FAIL:kprobe_test7_result unexpectedkprobe_test7_result: actual 0 != expected 1
...
kprobe_multi_test_run:FAIL:kretprobe_test7_result unexpectedkretprobe_test7_result: actual 0 != expected 1clang17 does not have this issue.
Further investigation shows that kernel func
bpf_fentry_test7(), used in the above tests, is inlined by the compiler although it is marked as noinline.int noinline bpf_fentry_test7(struct bpf_fentry_test_t *arg) { return (long)arg; }
Quoted 6 lines
It is known that for simple functions like the above (e.g. just returning a constant or an input argument), the clang compiler may still do inlining for a noinline function. Adding 'asm volatile ("")' in the beginning of the
bpf_fentry_test7()can prevent inlining.Signed-off-by Yonghong Song <[email protected]>
comparison")
...might help this land in stable trees too. Thanks!
Quoted 9 lines
--- net/bpf/test_run.c | 1 + 1 file changed, 1 insertion(+)net/bpf/test_run.c +0
-0
net/bpf/test_run.c struct bpf_fentry_test_t { · line 549
Quoted 7 lines
net/bpf/test_run.c +1
-0@@ -543,6 +543,7 @@ struct bpf_fentry_test_t { int noinline bpf_fentry_test7(struct bpf_fentry_test_t *arg) { + asmvolatile ("" );return (long )arg; }
Is there a risk bpf_fentry_test8/9 might get inlined too?
Yonghong Song
reviewerQuoted 40 lines · Alan Maguire
Quoted 39 lines · Yonghong Song
With latest clang18, I hit
test_progsfailures for the following test:#13/2 bpf_cookie/multi_kprobe_link_api:FAIL #13/3 bpf_cookie/multi_kprobe_attach_api:FAIL #13 bpf_cookie:FAIL #75 fentry_fexit:FAIL #76/1 fentry_test/fentry:FAIL #76 fentry_test:FAIL #80/1 fexit_test/fexit:FAIL #80 fexit_test:FAIL #110/1 kprobe_multi_test/skel_api:FAIL #110/2 kprobe_multi_test/link_api_addrs:FAIL #110/3 kprobe_multi_test/link_api_syms:FAIL #110/4 kprobe_multi_test/attach_api_pattern:FAIL #110/5 kprobe_multi_test/attach_api_addrs:FAIL #110/6 kprobe_multi_test/attach_api_syms:FAIL #110 kprobe_multi_test:FAILFor example, for #13/2, the error messages are
...
kprobe_multi_test_run:FAIL:kprobe_test7_result unexpectedkprobe_test7_result: actual 0 != expected 1
...
kprobe_multi_test_run:FAIL:kretprobe_test7_result unexpectedkretprobe_test7_result: actual 0 != expected 1clang17 does not have this issue.
Further investigation shows that kernel func
bpf_fentry_test7(), used in the above tests, is inlined by the compiler although it is marked as noinline.int noinline bpf_fentry_test7(struct bpf_fentry_test_t *arg) { return (long)arg; }It is known that for simple functions like the above (e.g. just returning a constant or an input argument), the clang compiler may still do inlining for a noinline function. Adding 'asm volatile ("")' in the beginning of the
bpf_fentry_test7()can prevent inlining.
Quoted 6 lines
Quoted 1 line
Signed-off-by Yonghong Song <[email protected]>Fixes d923021c2ce12 ("bpf: Add tests for PTR_TO_BTF_ID vs. nullcomparison")
...might help this land in stable trees too. Thanks!
I didn't put a Fixes tag since the issue is caused by upgrading clang compiler. The commit d923021c2ce12 does not have any issues at its patch-applying time.
Since this is not a core kernel functionality bug and it is due to upgrading clang compiler, I think backport is optional. But if maintainers think I should add Fixes tag, I can do it in the next revision.
Quoted 11 lines
Quoted 11 lines
--- net/bpf/test_run.c | 1 + 1 file changed, 1 insertion(+)net/bpf/test_run.c +0
-0@@ -543,6 +543,7 @@ struct bpf_fentry_test_t {
Quoted 7 lines
net/bpf/test_run.c struct bpf_fentry_test_t { · line 549
Quoted 6 lines
net/bpf/test_run.c +1
-0int noinline bpf_fentry_test7(struct bpf_fentry_test_t *arg) { + asmvolatile ("" );return (long )arg; }Is there a risk bpf_fentry_test8/9 might get inlined too?
So far, not. I cannot predict in the future. But if this happens with more complicated expressions got inlined for a non-inline function, I can discuss with upstream.
This patch was applied to bpf/bpf.git (master) by Daniel Borkmann <[email protected]>:
Hello:
This patch was applied to bpf/bpf.git (master)
by Daniel Borkmann <[email protected]>:
Quoted 17 lines · On Sat, 26 Aug 2023 13:08:43 -0700 you wrote:
With latest clang18, I hit
test_progsfailures for the following test:#13/2 bpf_cookie/multi_kprobe_link_api:FAIL #13/3 bpf_cookie/multi_kprobe_attach_api:FAIL #13 bpf_cookie:FAIL #75 fentry_fexit:FAIL #76/1 fentry_test/fentry:FAIL #76 fentry_test:FAIL #80/1 fexit_test/fexit:FAIL #80 fexit_test:FAIL #110/1 kprobe_multi_test/skel_api:FAIL #110/2 kprobe_multi_test/link_api_addrs:FAIL #110/3 kprobe_multi_test/link_api_syms:FAIL #110/4 kprobe_multi_test/attach_api_pattern:FAIL #110/5 kprobe_multi_test/attach_api_addrs:FAIL #110/6 kprobe_multi_test/attach_api_syms:FAIL #110 kprobe_multi_test:FAIL
Quoted 1 line
[...]
Here is the summary with links:
- [bpf-next] bpf: Prevent inlining of bpf_fentry_test7()
https://git.kernel.org/bpf/bpf/c/32337c0a2824
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html