[PATCH v8] tests/fcntl.c: add test for print_f_owner_ex

Zhibin Li 08826794brmt at gmail.com
Mon Apr 30 10:14:09 UTC 2018


Hi, all

I think my progress may be a little slow now considering I haven't made
this patch
acceptable yet. I submitted my proposal just before the deadline without
discussion so there
must be a lot of problems. In this community bonding period I'm trying to
make myself ready
for coding but I kinda don't know which specific part of tests I should
start with. The ratio of
coverage on codecov is a good reference. Should I just start with the part
with the least coverage?
I hope you guys can tell me something about what is feasible and what is
not.

Thanks,
Zhibin Li

On Mon, Apr 30, 2018 at 5:54 PM, Zhibin Li <08826794brmt at gmail.com> wrote:

> *tests/fcntl.c include <stdarg.h>
> (test_f_owner_ex_type_pid, test_f_owner_ex_umove,
> test_f_owner_ex): New functions.
> (main): Use test_f_owner_ex.
> ---
>  tests/fcntl.c | 53 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 53 insertions(+)
>
> diff --git a/tests/fcntl.c b/tests/fcntl.c
> index 4f62ca2a..c05d00a1 100644
> --- a/tests/fcntl.c
> +++ b/tests/fcntl.c
> @@ -69,12 +69,65 @@ test_flock64(void)
>  #endif
>  }
>
> +static long
> +test_f_owner_ex_type_pid(const int cmd, const char *const cmd_name,
> +                        const int type, const char *const type_name,
> +                        pid_t pid)
> +{
> +       TAIL_ALLOC_OBJECT_CONST_PTR(struct f_owner_ex, fo);
> +
> +       fo->type = type;
> +       fo->pid = pid;
> +       long rc = invoke_test_syscall(cmd, fo);
> +       printf("%s(0, %s, {type=%s, pid=%d}) = %s\n",
> +              TEST_SYSCALL_STR, cmd_name, type_name, fo->pid,
> sprintrc(rc));
> +
> +       long rc_efault = invoke_test_syscall(cmd, (void *) (uintptr_t) fo
> + 1);
> +       printf("%s(0, %s, %p) = %s\n",
> +              TEST_SYSCALL_STR, cmd_name, (void *) (uintptr_t) fo + 1,
> sprintrc(rc_efault));
> +
> +       return rc;
> +}
> +
> +static void
> +test_f_owner_ex_umove_or_printaddr(const int type, const char *const
> type_name, pid_t pid)
> +{
> +       long rc = test_f_owner_ex_type_pid(ARG_STR(F_SETOWN_EX),
> +                                          type, type_name, pid);
> +       if (!rc)
> +               test_f_owner_ex_type_pid(ARG_STR(F_GETOWN_EX),
> +                                        type, type_name, pid);
> +}
> +
> +static void
> +test_f_owner_ex(void)
> +{
> +       unsigned int i;
> +       unsigned int j;
> +       static const struct {
> +               int type;
> +               const char *type_name;
> +               pid_t pid[2];
> +       } a[] = {
> +               { ARG_STR(F_OWNER_TID), { 1234567890, 20 } },
> +               { ARG_STR(F_OWNER_PID), { 1298126790, 30 } },
> +               { ARG_STR(F_OWNER_PGRP), { 1294567890, 40 } }
> +       };
> +
> +       for (i = 0; i < ARRAY_SIZE(a); i++) {
> +               for (j = 0; j < ARRAY_SIZE(a[0].pid); j++) {
> +                       test_f_owner_ex_umove_or_printaddr(a[i].type,
> a[i].type_name, a[i].pid[j]);
> +               }
> +       }
> +}
> +
>  int
>  main(void)
>  {
>         create_sample();
>         test_flock();
>         test_flock64();
> +       test_f_owner_ex();
>
>         puts("+++ exited with 0 +++");
>         return 0;
> --
> 2.14.3
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.strace.io/pipermail/strace-devel/attachments/20180430/d4053f65/attachment.html>


More information about the Strace-devel mailing list