Re: [PATCH bpf 2/2] selftests/bpf: Add selftest for calling global functions from freplace

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

 



Andrii Nakryiko <andrii.nakryiko@xxxxxxxxx> writes:

> On Fri, Jun 3, 2022 at 8:42 AM Toke Høiland-Jørgensen <toke@xxxxxxxxxx> wrote:
>>
>> Add a selftest that calls a global function with a context object parameter
>> from an freplace function to check that the program context type is
>> correctly converted to the freplace target when fetching the context type
>> from the kernel BTF.
>>
>> Signed-off-by: Toke Høiland-Jørgensen <toke@xxxxxxxxxx>
>> ---
>>  .../selftests/bpf/prog_tests/fexit_bpf2bpf.c  | 13 ++++++++++
>>  .../bpf/progs/freplace_global_func.c          | 24 +++++++++++++++++++
>>  2 files changed, 37 insertions(+)
>>  create mode 100644 tools/testing/selftests/bpf/progs/freplace_global_func.c
>>
>> diff --git a/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c b/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c
>> index d9aad15e0d24..6e86a1d92e97 100644
>> --- a/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c
>> +++ b/tools/testing/selftests/bpf/prog_tests/fexit_bpf2bpf.c
>> @@ -395,6 +395,17 @@ static void test_func_map_prog_compatibility(void)
>>                                      "./test_attach_probe.o");
>>  }
>>
>> +static void test_func_replace_global_func(void)
>> +{
>> +       const char *prog_name[] = {
>> +               "freplace/test_pkt_access",
>> +       };
>
> empty line between variables and statements
>
>> +       test_fexit_bpf2bpf_common("./freplace_global_func.o",
>> +                                 "./test_pkt_access.o",
>> +                                 ARRAY_SIZE(prog_name),
>> +                                 prog_name, false, NULL);
>> +}
>> +
>>  /* NOTE: affect other tests, must run in serial mode */
>>  void serial_test_fexit_bpf2bpf(void)
>>  {
>> @@ -416,4 +427,6 @@ void serial_test_fexit_bpf2bpf(void)
>>                 test_func_replace_multi();
>>         if (test__start_subtest("fmod_ret_freplace"))
>>                 test_fmod_ret_freplace();
>> +       if (test__start_subtest("func_replace_global_func"))
>> +               test_func_replace_global_func();
>>  }
>> diff --git a/tools/testing/selftests/bpf/progs/freplace_global_func.c b/tools/testing/selftests/bpf/progs/freplace_global_func.c
>> new file mode 100644
>> index 000000000000..d9f8276229cc
>> --- /dev/null
>> +++ b/tools/testing/selftests/bpf/progs/freplace_global_func.c
>> @@ -0,0 +1,24 @@
>> +// SPDX-License-Identifier: GPL-2.0
>> +/* Copyright (c) 2019 Facebook */
>> +#include <linux/stddef.h>
>> +#include <linux/bpf.h>
>> +#include <bpf/bpf_helpers.h>
>> +#include <bpf/bpf_endian.h>
>> +#include <bpf/bpf_tracing.h>
>> +
>> +__attribute__ ((noinline))
>
> __noinline
>
>> +int test_ctx_global_func(struct __sk_buff *skb)
>> +{
>> +       volatile int retval = 1;
>> +       return retval;
>
> just curious, why volatile instead of direct `return 1;`? Also, empty
> line between variable and return.

To prevent the compiler from optimising away the whole thing. I was
actually under the impression that it shouldn't do that for global
functions, but, well, it does and the 'volatile' keeps it from doing
so...

Will fix the other nits and resubmit.

-Toke





[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