[PATCH] selftests/exec: Check if the syscall exists and bail if not

[PATCH] selftests/exec: Check if the syscall exists and bail if not

[Michael Ellerman]

On systems which don't implement sys_execveat(), this test produces a
lot of output.

Add a check at the beginning to see if the syscall is present, and if
not just note one error and return.

Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
---
 tools/testing/selftests/exec/execveat.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/tools/testing/selftests/exec/execveat.c b/tools/testing/selftests/exec/execveat.c
index e238c9559caf..b87e4a843bea 100644
--- a/tools/testing/selftests/exec/execveat.c
+++ b/tools/testing/selftests/exec/execveat.c
@@ -234,6 +234,14 @@ static int run_tests(void)
 	int fd_cloexec = open_or_die("execveat", O_RDONLY|O_CLOEXEC);
 	int fd_script_cloexec = open_or_die("script", O_RDONLY|O_CLOEXEC);
 
+	/* Check if we have execveat at all, and bail early if not */
+	errno = 0;
+	execveat_(-1, NULL, NULL, NULL, 0);
+	if (errno == -ENOSYS) {
+		printf("[FAIL] ENOSYS calling execveat - no kernel support?\n");
+		return 1;
+	}
+
 	/* Change file position to confirm it doesn't affect anything */
 	lseek(fd, 10, SEEK_SET);
 
-- 
2.1.0

Re: [PATCH] selftests/exec: Check if the syscall exists and bail if not

[David Drysdale]

On Wed, Jan 21, 2015 at 7:41 AM, Michael Ellerman <mpe@ellerman.id.au> wrote:
> On systems which don't implement sys_execveat(), this test produces a
> lot of output.
>
> Add a check at the beginning to see if the syscall is present, and if
> not just note one error and return.

Good point, thanks.

> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
> ---
>  tools/testing/selftests/exec/execveat.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
>
> diff --git a/tools/testing/selftests/exec/execveat.c b/tools/testing/selftests/exec/execveat.c
> index e238c9559caf..b87e4a843bea 100644
> --- a/tools/testing/selftests/exec/execveat.c
> +++ b/tools/testing/selftests/exec/execveat.c
> @@ -234,6 +234,14 @@ static int run_tests(void)
>         int fd_cloexec = open_or_die("execveat", O_RDONLY|O_CLOEXEC);
>         int fd_script_cloexec = open_or_die("script", O_RDONLY|O_CLOEXEC);
>
> +       /* Check if we have execveat at all, and bail early if not */
> +       errno = 0;
> +       execveat_(-1, NULL, NULL, NULL, 0);
> +       if (errno == -ENOSYS) {

Could we change this to ENOSYS (no minus) and also change
the execveat_() function similarly, so that a binary built where
__NR_execveat is available but running where it isn't also exits
early?  (My bad for having the minus sign in execveat_() in the
first place -- fingers too used to kernel mode.)

Thanks!

> +               printf("[FAIL] ENOSYS calling execveat - no kernel support?\n");
> +               return 1;
> +       }
> +
>         /* Change file position to confirm it doesn't affect anything */
>         lseek(fd, 10, SEEK_SET);
>
> --
> 2.1.0
>

Re: [PATCH] selftests/exec: Check if the syscall exists and bail if not

[Michael Ellerman]

On Wed, 2015-01-21 at 10:22 +0000, David Drysdale wrote:
> On Wed, Jan 21, 2015 at 7:41 AM, Michael Ellerman <mpe@ellerman.id.au> wrote:
> > On systems which don't implement sys_execveat(), this test produces a
> > lot of output.
> >
> > Add a check at the beginning to see if the syscall is present, and if
> > not just note one error and return.
> 
> Good point, thanks.
> 
> > diff --git a/tools/testing/selftests/exec/execveat.c b/tools/testing/selftests/exec/execveat.c
> > index e238c9559caf..b87e4a843bea 100644
> > --- a/tools/testing/selftests/exec/execveat.c
> > +++ b/tools/testing/selftests/exec/execveat.c
> > @@ -234,6 +234,14 @@ static int run_tests(void)
> >         int fd_cloexec = open_or_die("execveat", O_RDONLY|O_CLOEXEC);
> >         int fd_script_cloexec = open_or_die("script", O_RDONLY|O_CLOEXEC);
> >
> > +       /* Check if we have execveat at all, and bail early if not */
> > +       errno = 0;
> > +       execveat_(-1, NULL, NULL, NULL, 0);
> > +       if (errno == -ENOSYS) {
> 
> Could we change this to ENOSYS (no minus) and also change
> the execveat_() function similarly, so that a binary built where
> __NR_execveat is available but running where it isn't also exits
> early?  (My bad for having the minus sign in execveat_() in the
> first place -- fingers too used to kernel mode.)

Ah yeah, me too, -ENOSYS just came naturally.

Will fix up and retest and resend.

cheers