Re: [PATCH bpf-next v3 6/7] selftests/bpf: Add struct argument tests with fentry/fexit programs.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, Aug 30, 2022 at 12:12:08AM +0200, Daniel Borkmann wrote:
> On 8/28/22 4:55 AM, Yonghong Song wrote:
> > Add various struct argument tests with fentry/fexit programs.
> > Also add one test with a kernel func which does not have any
> > argument to test BPF_PROG2 macro in such situation.
> > 
> > Signed-off-by: Yonghong Song <yhs@xxxxxx>
> > ---
> >   .../selftests/bpf/bpf_testmod/bpf_testmod.c   |  48 ++++++++
> >   .../selftests/bpf/prog_tests/tracing_struct.c |  63 ++++++++++
> >   .../selftests/bpf/progs/tracing_struct.c      | 114 ++++++++++++++++++
> >   3 files changed, 225 insertions(+)
> >   create mode 100644 tools/testing/selftests/bpf/prog_tests/tracing_struct.c
> >   create mode 100644 tools/testing/selftests/bpf/progs/tracing_struct.c
> > 
> 
> For s390x these tests need to be deny-listed due to missing trampoline support..
> 
>   All error logs:
>   test_fentry:PASS:tracing_struct__open_and_load 0 nsec
>   libbpf: prog 'test_struct_arg_1': failed to attach: ERROR: strerror_r(-524)=22
>   libbpf: prog 'test_struct_arg_1': failed to auto-attach: -524
>   test_fentry:FAIL:tracing_struct__attach unexpected error: -524 (errno 524)
>   #209     tracing_struct:FAIL
>   Summary: 189/972 PASSED, 27 SKIPPED, 1 FAILED
> 
> However, looks like the no_alu32 ones on x86 fail:
> 
>   [...]
>   #207     trace_printk:OK
>   #208     trace_vprintk:OK
>   test_fentry:PASS:tracing_struct__open_and_load 0 nsec
>   test_fentry:PASS:tracing_struct__attach 0 nsec
>   trigger_module_test_read:PASS:testmod_file_open 0 nsec
>   test_fentry:PASS:trigger_read 0 nsec
>   test_fentry:PASS:t1:a.a 0 nsec
>   test_fentry:PASS:t1:a.b 0 nsec
>   test_fentry:PASS:t1:b 0 nsec
>   test_fentry:PASS:t1:c 0 nsec
>   test_fentry:PASS:t1 nregs 0 nsec
>   test_fentry:PASS:t1 reg0 0 nsec
>   test_fentry:PASS:t1 reg1 0 nsec
>   test_fentry:FAIL:t1 reg2 unexpected t1 reg2: actual 7327499336969879553 != expected 1

I'm getting the same, I think it's because the argument is int (4 bytes)
while the register is 8, we need to cast to int before we check for the
argument value

jirka

>   test_fentry:PASS:t1 reg3 0 nsec
>   test_fentry:PASS:t1 ret 0 nsec
>   test_fentry:PASS:t2:a 0 nsec
>   test_fentry:PASS:t2:b.a 0 nsec
>   test_fentry:PASS:t2:b.b 0 nsec
>   test_fentry:PASS:t2:c 0 nsec
>   test_fentry:PASS:t2 ret 0 nsec
>   test_fentry:PASS:t3:a 0 nsec
>   test_fentry:PASS:t3:b 0 nsec
>   test_fentry:PASS:t3:c.a 0 nsec
>   test_fentry:PASS:t3:c.b 0 nsec
>   test_fentry:PASS:t3 ret 0 nsec
>   test_fentry:PASS:t4:a.a 0 nsec
>   test_fentry:PASS:t4:b 0 nsec
>   test_fentry:PASS:t4:c 0 nsec
>   test_fentry:PASS:t4:d 0 nsec
>   test_fentry:PASS:t4:e.a 0 nsec
>   test_fentry:PASS:t4:e.b 0 nsec
>   test_fentry:PASS:t4 ret 0 nsec
>   test_fentry:PASS:t5 ret 0 nsec
>   #209     tracing_struct:FAIL
>   #210     trampoline_count:OK
>   [...]



[Index of Archives]     [Linux Samsung SoC]     [Linux Rockchip SoC]     [Linux Actions SoC]     [Linux for Synopsys ARC Processors]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]


  Powered by Linux